Philosophical Foundations of Neuroscience 9781405108386

Table of contents :
Preface.

Part I: Philosophical Problems In Neuroscience: Their Historical and Conceptual Roots:.

1. The Early Growth Of Neuroscientific Knowledge: The Integrative Action Of The Nervous System.

Aristotle, Galen And Nemesius: The Origins Of The Ventricular Doctrine.

Fernel And Descartes: The Demise Of The Ventricular Doctrine.

The Cortical Doctrine Of Willis And Its Aftermath.

The Conception Of A Reflex: Bell, Magendie And Marshall Hall.

Localizing Function In The Cortex: Broca, Fritz And Hitzig.

The Integrative Action Of The Nervous System: Sherrington.

2. The Cortex And The Mind In The Work Of Sherrington And His Proteges.

Charles Sherrington: The Continuing Cartesian Impact.

Edgar Adrian: Hesitant Cartesianism.

John Eccles And The ‘Liaison Brain’.

Wilder Penfield And The ‘Highest Brain Mechanism’.

3. The Mereological Fallacy And Its Manifestation In Contemporary Neuroscientific Thought.

Mereological Confusions In Cognitive Neuroscience: (Crick, Edelman, Blakemore, Young, Frisby, Gregory, Marr, Johnson-Laird).

Methodological Qualms: (Ullman, P.S. Churchland, Blakemore, Zeki, Young, Milner Squire And Kandel, Marr, Frisby, Sperry).

On The Grounds For Ascribing Psychological Predicates To A Being: (Crick, Baars).

On The Grounds For Misascribing Psychological Predicates To An Inner Entity: (Damasio, Edelman And Tononi, Kosslyn And Ochsner, Searle, James, Libet, Humphrey, Blakemore, Crick).

The Inner: (Damasio).

Introspection: (Humphrey, Johnson-Laird, Weiskrantz).

Privileged Access: Direct And Indirect: (Blakemore).

Privacy Or Subjectivity: (Searle).

The Meaning Of Psychological Predicates And How They Are Learnt: (Searle).

Of The Mind And Its Nature: (Gazzaniga, Doty).

Part II: Human Faculties and Contemporary Neuroscience: an Analysis:.

Preliminaries.

Brain-Body Dualism: (Kandel Schwartz And Jessell, Libet).

The Project: (Gazzaniga).

The Category Of The Psychological: (Nagel, P.M. Churchland And P.S. Churchland).

4. Sensation and Perception.

Sensation: (Searle, Libet, Geldard And Sherrick).

Perception: (Ledoux, Crick).

Perception As The Causation Of Sensations: Primary And Secondary Qualities: (Kandel Schwartz And Jessell, Rock).

Perception As Hypothesis Formation: Helmholtz: (Helmholtz, Gregory, Glynn, Young).

Citation preview

PHILOSOPHICAL FOUNDATIONS OF NEUROSCIENCE

M. R . Bennett and E M. S. Hacker

Blackwefl Publishing

Reviews of Philosophical Foundations of Neuroscience: 'This remarkable book, the product of a collaboration between a philosopher and tieuroscientist, shows that the claims made on behalf of cognitive science are ill-founded. The real significance of impressive recent developments in the study of the brain, they allege, has been clouded by philosophical confusion in the way in which these results have been presented. The authors document their complaint in a clear and patient manner. . . . They disentangle the confusions by setting out clearly the contrasting but complementary roles of philosophy and neuroscience in this area. The book will certainly arouse opposition. . . . But if it causes controversy, it is controversy that is long overdue. It is to be hoped that it will be widely read among those in many different disciplines who are interested in the brain and the mind.' Sir Anthony Kenny, President of the British Academy (1989-Í993) 'Overall the book provides the most thorough critical survey of the ruling theories of mental phenomena as they figure in contemporary science. The attention to detail is meticulous, and the philosophical analysis outstandingly lucid. Contemporary scientists and philosophers may not like Bennett and Hacker's conclusions, but they will hardly be able to ignore them. The work is a formidable achievement.' John Cottingham, Professor of Philosophy, University of Reading 'Contemporary neuroscience is an exciting, ebullient field and its practitioners are not much given to self-doubt. This dissection of the field by Bennett and Hacker ought to provoke some misgivings. Arguing for a sharp distinction between conceptual analysis of our everyday psychological concepts on the one hand and empirical, neuroscientific investigation on the other, Bennett and Hacker conclude that many neuroscientists - and some of their philosopher friends — have ignored or muddied that distinction at their peril. In particular, they argue that the misuse of psychological concepts in the interpretation of neural processes does not lead to testable or even false claims, but to nonsense. Neuroscientists, psychologists and philosophers will be challenged - and educated — by this sustained and well-informed critique.' Paul L. Harris, Professor, Human Development and Psychology, Harvard University '[It] will certainly, for a long time to come, be the most important contribution to the mind-body problem there is.' G. H. von Wright (1916-2003), Research Professor, Academy of Finland and Professor of Philosophy at Cambridge, Cornell and Helsinki 'Sweeping, argumentative, and brilliant, this book will provoke widespread discussion among philosophers and neuroscientists alike.' Dennis Patterson, Notre Dame Philosophical Review 'Devastating critiques of psychologists and neuroscientists. . . . Whether this book leads to a reconfiguring of contemporary neuroscience and the philosophy associated with it will tell us much about the dynamics of contemporary intellectual life.' Anthony O'Hear, Philosophy

'This book is a joy to read. . . . A model of clarity and directedness . . . [Bennett and Hacker] have produced that rarity of scholarship, a genuinely interdisciplinary work that succeeds. . . . This is a wonderful book that will illuminate, provoke and delight professional scientists, philosophers, and general readers alike.' Damián Grace, Australian Book Review 'Clinical precision and . . . relentless good sense . . . [a] thoughtful and useful treatise.' Daniel N. Robinson, Philosophy 'Mandatory reading for anybody interested in neuroscience and consciousness research. The vast spectrum of material in philosophy and neuroscience that Bennett and Hacker consider is impressive and their discussion is thorough and illuminating.' Axel Kohler, Human Nature Review 'A delicious cake of a book in which Bennett and Hacker guide the reader through a conceptual minefield of confusions repeatedly made by neuroscientists and philosophers alike.'* Constantine Sandis, Metapsychology 'Anyone who has ever framed a theory or explained one should read this book - at the risk of forever falling silent.' The Rector, University of Sydney, Obiter Dicta 'Impressively lucid . . . Bennett and Hacker unquestionably succeed in challenging our own concepts, examine them for dross, and strive to home in on fundamentals.' Neil Spurway, Journal of the European Society for Study of Science and Theology 'The fruit of a unique co-operation between a neuroscientist and a philosopher. . . . An excellent book that should be read by all philosophers of cognition and all researchers in the cognitive neurosciences.' Herman Philipse, ABG #2, De Academische Bockengids '[Tjhere are, I think, grounds for hope that this book will do an enormous amount of good, both in correcting philosophical confusion within neuroscience and in promoting a new style of dialogue between neuroscience and philosophy.' David Cockburn, Philosophical investigations

© 2003 by M. R. Bennett and P. M. S. Hacker BLACKWELL PUBLISHING 350 Main Street, Maiden, MA 02148-5020, USA 9600 Garsington Road, Oxford OX4 2DQ, UK 550 Swanston Street, Carlton, Victoria 3053, Australia The right of M. R. Bennett and P. M. S. Hacker to be identified as the Authors of this Work has been asserted in accordance with the UK Copyright, Designs, and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act 1988, without the prior permission of the publisher. First published 2003 by Blackwell Publishing Ltd 10

2010

Library of Congress Cataloging-in-Publication Data Bennett, M. R. Philosophical foundations of neuroscience / M. R. Bennett and P. M. S. Hacker. p. cm. Includes bibliographical references and index. ISBN 1-4051-0855-X (hbk. : alk. paper) - ISBN 1-4051-0838-X (pbk, : alk. paper) 1. Cognitive neuroscience—Philosophy. I. Hacker, P. M. S. (Peter Michael Stephan) II. Title QP360.5 .B4Ó5 2003 612.8'2'01-dc21 2002028110 ISBN 978-1-4051-0855-3 (hbk. : alk. paper) - ISBN 978-1-4051-0838-6 (pbk. : alk. paper) A catalogue record for this title is available from the British Library. Set in 10/12 pt Bembo by Graphicraft Ltd, Hong Kong Printed and bound in Malaysia by KHL Printing Co Sdn Bhd The publisher's policy is to use permanent paper from milis that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com

For Gillian and Jocelyn

Contents

Foreword Acknowledgements Introduction Part I

1

T h e Early Growth of Neuroscientific Knowledge: T h e Integrative Action of the Nervous System 1.1 1.2 1.3 1.4 1.5 1.6

2

3

Philosophical Problems in Neuroscience: T h e i r Historical and Conceptual R o o t s

Aristotle, Galen and Nemesius: The Origins of the Ventricular Doctrine Femel and Descartes: The Demise of the Ventricular Doctrine The Cortical Doctrine of Willis and its Aftermath The Concept of a Reflex: Bell, Magendie and Marshall Hall Localizing Function in the Cortex: Broca, Fritsch and Hitzig The Integrative Action of the Nervous System: Sherrington

xiii xvii 1

9

11

12 23 30 33 38 41

T h e Cortex and the Mind in the W o r k of Sherrington and his Proteges

43

2.1 2.2 2.3 2.4

43 47 49 57

Charles Sherrington: T h e Continuing Cartesian Impact Edgar Adrian: Hesitant Cartesianism John Eccles and the 'Liaison Brain' Wilder Penfield and the 'Highest Brain Mechanism'

T h e Mereological Fallacy in Neuroscience

68

3.1

68

Mereological Confusions in Cognitive Neuroscience (Crick, Edelman, Blakemore, Young, Damasio, Frisby, Gregory, Man, Johnson-Laird)

Contents

vm 3.2

3.3 3.4

3.5 3.6 3.7 3.8 3.9 3.10

Part II

Methodological Qualms (Ulltnan, Blakemore, Zeki, Young, Milner, Squire and Kandel, Man, Frisby, Speriy) O n the Grounds for Ascribing Psychological Predicates to a Being O n the Grounds for Misascribing Psychological Predicates to an Inner Entity (Damasio, Edelman and Tononi, Kosslyn and Ochsner, Searle, James, Libet, Humphrey, Blakemore, Crick) The Inner (Damasio) Introspection (Humphrey, Johnson-Laird, Weiskrantz) Privileged Access: Direct and Indirect (Blakemore) Privacy or Subjectivity (Searle) The Meaning of Psychological Predicates and How they are Learnt Of the Mind and its Nature (Gazzaniga, Doty)

Human Faculties and Contemporary Neuroscience: An Analysis

74

81 85

88 90 92 94 97 103

109

Preliminaries

111

1 2 3

111 114 117

4

Brain-Body Dualism The Project The Category of the Psychological

Sensation and Perception

121

4.1

121

4.2 4.2.1

4.2.2 4.2.3

4.2.4

Sensation (Searle, Libet, Geldard and Sherrick) Perception (Crick) Perception as the causation of sensations: primary and secondary qualities (Kandel, Schwartz and Jessell, Rock) Perception as hypothesis formation: Helmholtz (Helrnholtz, Gregory, Glynn, Young) Visual images and the binding problem (Sherrington, Damasio, Edelman, Crick, Kandel and Wurtz, Gray and Singer, Barlow) Perception as information processing: Marr's theory of vision (Man, Frisby, Crick, Ullman)

125

128 135 137

143

5

The Cognitive Powers

148

5.1 5.1.1 5.1.2

148 149 151

Knowledge and its Kinship with Ability Being able to and Knowing how to Possessing Knowledge and Containing Knowledge (LeDoux, Young, Zeki, Blakemore, Crick, Gazzaniga) 5.2 Memory (Milner Squire and Kandel) 5.2.1 Declarative and non-declarative memory (Milner, Squire and Kandel) 5.2.2 Storage, retention and memory traces (LeDoux, Squire and Kandel; Gazzaniga, Mangun and Ivry; James, Kohler, Glynn;. Bennett Gibson, and Robinson; Damasio) 6

158

172

6.1

172

6.3.1

Belief (Crick) Thinking Imagination and Mental Images (Blakemore, Posner and Raichle, Shepard) The logical features of mental imagery (Galton, Richardson, Kosslyn and Ochsner, Finke, Luria, Shepard, Meudell, Betts, Marks, Shepard and Metzler, Cooper and Shepard, Posner and Raichle)

175 180 187

Emotion

199

7.1

199

7.2 7.2.1 7.2.2 8

155

T h e Cogitative Powers

6.2 6.3

1

154

Affections (Rolls, Damasio) The Emotions: A Preliminary Analytical Survey Neuroscientists' confusions (LeDoux, Damasio, James) Analysis of the emotions

203 207 216

Volition and Voluntary M o v e m e n t

224

8.1 8.2

224 228

8.3

Volition Libet's Theory of Voluntary Movement (Libet, Frith et al.) Taking Stock

231

Contents

X

Part III

9

Consciousness and Contemporary Neuroscience: An Analysis

237

Intransitive and Transitive Consciousness

239

9.1

239

9.2 9.3 9.4

10

Consciousness and the Brain (Albright, Jessell, Kandel and Posner, Edelman and To'noni; Glynn, Greenfield, Llinás, Gazzaniga, Searle, Johnson-Laird, Chalmers, Dennett, Gregory, Crick and Koch, Frisby) Intransitive Consciousness (Searle, Dennett) Transitive Consciousness and its Forms Transitive Consciousness: A Partial Analysis

248 253

Conscious Experience, Mental States and Qualia

261

10.1

261

10.2 10.2.1

10.3

10.3.1 10.3.2 10.3.3 10.3.4 10.3.5

11

244

Extending the Concept of Consciousness (Libet, Baars, Crick, Edelman, Searle, Chalmers) Conscious Experience and Conscious Mental States Confusions regarding unconscious behef and unconscious activities of the brain (Searle, Baars) Qualia (Searle, Chalmers, Glynn, Damasio, Edelman and Tononi, Nagel, Dennett) 'How it feels' to have an experience (Searle, Edelman and Tononi, Chalmers) Of there being something which it is like . . . (Nagel) The qualitative character of experience Thises and fftwses (Chalmers, Crick) Of the communicability and describability of qualia (Nagel, Edelman, Glynn, Sperry)

263 268 271

274 Til 281 282 284

Puzzles about Consciousness

293

11.1 11.2

293

11.3

A Budget of Puzzles O n Reconciling Consciousness or Subjectivity with our Conception of an Objective Reality (Searle, Chalmers, Dennett, Penrose) On the Question of how Physical Processes can give rise to Conscious Experience (Huxley, Tyndall, Humphrey, Glynn, Edelman, Damasio)

294

302

Contents

11.4

Of the Evolutionary Value of Consciousness (Chalmers, Barlow, Penrose, Humphrey, Searle) The Problem of Awareness (Johnson-Laird, Blakemore) Other Minds and Other Animals (Crick, Edelman, Weiskmntz, Baars)

11.5 11.6

12

314 316

323

12.1 12.2 12.3

323 324

Self-Consciousness and the Self Historical Stage Setting: Descartes, Locke, Hume and James Current Scientific and Neuroscientific Reflections on the Nature of Self-Consciousness • (Damasio, Edelman, Humphrey, Blakemore, Johnson-Laird) The Illusion of a ' S e l f (Damasio, Humphrey, Blakemore) The Horizon of Thought, Will and Affection Thought and language (Damasio, Edelman and Tononi, Galton, Penrose) Self-Consciousness (Edelman, Penrose)

12.5 12.5.1 12.6

Part IV

On Method

328 331 334 337 346

353

Reductionism

355

13.1

355

13.2 13.2.1 13.2.2

13.2.3 13.2.4

14

307

Self-Consciousness

VIA

13

XI

Ontological and Explanatory Reductionism (Crick, Blakemore) Reduction by Elimination (P. M. and P. S. Churchland) Are our ordinary psychological concepts theoretical? (P. M. Churchland) Are everyday generalizations about human psychology laws of a theory? (P. M. Churchland) Eliminating all that is human (P. M. and P. S. Churchland, Dawkins) Sawing off the branch on which one sits

366 367

370 372 376

Methodological Reflections

378

14.1

379

14.2

Linguistic Inertia and Conceptual Innovation (P. S. Churchland) The 'Poverty of English' Argument

386

Contents

XII

14.3

14.3.1 14.4 14.4.1 14.4.2 14.5

From Nonsense to Sense: The Proper Description of the Results of Commissurotomy (Crick, Sperry, Gazzaniga, Wolford Miller and Gazzaniga, Doty) The case of blind-sight: misdescription and illusory explanation (Weiskrantz) Philosophy and Neuroscience (Glynn, Edelman, Edelman and Tononi, Crick, Zeki) What philosophy can and what it cannot do What neuroscience can and what it cannot do (Crick, Edelman, Zeki) Why it Matters

388 393 396 399 405 408

Appendices

411

Appendix 1 Daniel Dennett

413

1 2 3 4

415 419 427 431

Dennett's Methodology and Presuppositions The Intentional Stance Heterophenomenological Method Consciousness

Appendix 2 John Searle

436

1 2 3

436 443 449

Philosophy and Science Searle's Philosophy of Mind The Traditional Mind-Body Problem

Index

453

Foreword

This book was simply waiting to be written. The reductionist agenda in biological science has generated so many conceptual difficulties that someone, sometime, had to analyse these problems in depth from outside the reductionist viewpoint. That a neurophysiologist and a philosopher should combine to do so is also a sign of the times. As biology moves on to address the complexity and extraordinary subtlety of life, now that it has broken it down into its smallest pieces, we will find this kind of combination of skills and ways of thinking even more necessary. As the authors make clear, philosophy (at least in the analytical form practised here) and empirical science are not in opposition. Rather they deal with different kinds of question. Yet, since a conceptual scheme is necessary to any fruitful experimentation, we cannot avoid asking both kinds. Keeping a clear head while we do so is not as easy as it may seem! I must issue a warning: this book is highly controversial. Some of my scientific colleagues will strongly challenge, and will surely be deeply provoked by, the claim that neuroscience has frequently and systematically confused conceptual and empirical questions. To them I would say, first, that the authors clearly recognize the brilliance and phenomenal achievements of the scientists whose conceptual work they analyse. This is emphatically not a book debunking experimental science, any more than the fact that most physiologists now dismiss the dualist philosophy of Sherrington or Eccles detracts in any way from recognizing the immense significance of their scientific achievements. We find it perfectly possible to admire the experimental and associated analytical skills while wincing when we see how completely trapped they were in their outdated and indefensible philosophical position. Second, I would appeal for some patience and humility. Patience, because as a physiologist who has interacted with (and published with) professional philosophers of various persuasions for over 40 years,1 I have to say that I find scientists unthinkingly debunking

1

See, for example, Noble, D. (1989), What Do Intentions Do? In: Goals, No Goah and Own Goals, eds Alan Montefiore and Denis Noble, London: Unwin Hyman; and Noble, D. (1991), Biological Explanation and Intentional Behaviour. In: Modelling the Mind, eds W. H. Newton-Smith and K. Wilkes, Oxford: Oxford University Press.

XIV

Foreword

philosophy more often than the other way round. Humility, because the issues are of the utmost social importance. Some of the claims of reductionist science are not only conceptually incorrect or even unintelligible, they have major social implications. The words we use, the concepts by which we analyse and present biological discovery, deeply affect the way in which we see ourselves as human beings. For that reason, if for no other, a critical debate is necessary. The authors of this book have thrown down a major challenge in that debate. The controversial nature of this book arises because the particular reductionist philosophical position it criticizes is very widely held today within the scientific community (and also by some well-known philosophers). Moreover, for most of them, this position is a methodological necessity, perceived to be the only paradigm for science to successfully explain things. The first reaction to the counter-argument, as presented here, will be to protest that somehow science is being (unnecessarily?) circumscribed; that some problems are, as it were, being taken from its grasp. I would argue the other way. The first step to scientific progress is to ask the right questions. If we are conceptually confused, we will ask the wrong questions. The authors illustrate this in detail with many examples. It is hard to escape the confines and confusions of the culture in which one finds oneself. The history of philosophy, shows that, just as much, as the history of science. The central appeal of this book is to throw off the remaining legacy of the Cartesian confusions, first expressed as a duality of mind and body, but latterly expressed as a duality of brain and body. The authors show that, although the first required belief in a non-material substance, while the latter is wholly materialistic, many of the conceptual problems (essentially those of the 'ghost in the machine') are the same. For our dualist predecessors the ghost was an actual immaterial substance, for us it is 'the " I " ' (or 'inner eye' or whatever) that 'sees' the qualia that 'form our experience'. This is what may lead us to ask which group of cells, or even which neurone(!), is doing the 'seeing'. The point here is that simply replacing T or 'inner eye' by the brain or a part of the brain doesn't avoid the problem. The key to understanding the confusions here lies in an analysis of the logical conditions for ascribing mental and psychological properties. This is not easy. It involves one of the most difficult of twentieth-century philosophical ideas, that of the 'private language argument': what it is to say things like 'I feel pain' or 'I see red'. I struggled through the ramifications of this argument many years ago before writing my own contributions to the philosophy of biology. I wish I had had the benefit of the relatively easy path that Bennett and Hacker have provided. Even those who fundamentally disagree with their arguments (and I look forward to seeing them engage in debate) must surely acknowledge that this is a sustained and valuable exposition of an important and influential philosophical position. Although I would describe that position as philosophically radical (in the correct sense of that word: going back to basic roots and eradicating those that shouldn't be there), it is often dismissed by scientists as conservative because it may appear to restrict using language in new ways. Yet, they would argüe, science cannot advance without doing that. And what better way to achieve it than to start with metaphor or jacons de parley, consolidate with dead metaphor (metaphors that become part of everyday language — constructivists argue that that is the way language evolved) and finaDy end up with a change in our

Foreword

xv

conceptual scheme? Indeed, why not, if that is what will enlighten us, lead us into new conceptual territory, formulate new theories. But there is a simple test for whether that could work in any particular case. For each such metaphorical {or similar) change in use or meaning, or novel piece of terminology (such as 'qualia' or 'memes'), imagine stating its opposite, and then ask whether any conceivable experiment could test empirically between the two. The deep problem for many 'novel' concepts and language uses in reductionist approaches is that this test totally fails. The novel use of language is then not so much a scientific as a political or social tool. If you doubt this, try imagining an experiment to test between the existence or non-existence of qualia. Or for whether or not the brain makes representative maps (which are not homunculi incidentally). Or for brain states that 'explain' rational thought (rather than being a necessary physical basis for its existence). Surely we should only introduce new terminology where, as with quarks and black holes, we provide the empirical criteria for determining their existence? Perhaps the problem for many scientists is to imagine what would happen if we abandoned the universality of the reductionist approach. For sure, the nature of science would change. But so it should! W e would have to recognize that causation and explanation do not always run upwards from lower to higher levels. And, surely, at a time when we have already come to understand the extent to which causation runs in the opposite direction (higher-level states in biological systems even influence something as fundamentally lower-level as gene expression), how can we possibly imagine that we will progress without recognizing the validity of explanations at all levels? One of the criteria for determining the level at which explanation succeeds is to ask what can sensibly be ascribed at different levels. It does not make sense to look for explanations at levels lower than that for the applicability of the relevant predicates. This is particularly true of rational behaviour, including the use of language. The argument is basically very simple. We cannot, coherendy, deny our own rationality. Otherwise we would have difficulty meaning what we say or being convincing in saying it, which is precisely what happens in the sad cases of those mentally ill people who nevertheless are aware of, but can't help, their irrationality. If we really could succeed in 'reducing' rational behaviour simply to molecular or cellular causation then we would no longer be able meaningfully to express the truth of what we had succeeded in doing. But, thankfully, no such reduction is conceivable. We know what it is to be rational, and what it is to lose that capacity. That knowledge has nothing to do with the question whether there exist specific and causally sufficient neural states and interactions while I am writing this review, for example. Of course they do. And, if we can discover them, they may well provide a complete explanation for the mechanisms of my brain while thinking and writing. The main claim of anti-reductionism in science is that such a complete explanation of mechanisms at one level does not necessarily explain what exists and happens at higher levels. Indeed we may need to know about the higher levels in order to explain the lower-level data that form an input to the mechanisms involved (which is what must be the case in writing this review! - one of the inputs was my reading this book, but the book is not thereby 'inside' my brain). The most spectacular case of this need for higher-level understanding in modern biology is, of course, the genome, whose sequences will only be understood eventually in

XVI

Foreword

terms of higher-level function (genes don't come with functional names attached — nor do neurones!). I started my life in physiological research as a fully paid-up member of the reductionist club. In the 1960s you couldn't get much more reductionist than to discover ion channels in excitable cells and then to simulate their activity in a bottom-up approach. I did for heart cells what Hodgkin & Huxley did for nerve cells. It is through trying to extend this approach to higher physiological levels that I have come to see the conceptual and computational problems that arise in practice. 1 have written elsewhere on the impossibility of a completely bottom-up reconstruction of living systems.2 Concurrently I also interacted extensively with professional philosophers (of different schools of thought — including those who would support the reductionist agenda). Coming to adopt an integrationist agenda was not an easy road, either scientifically or philosophically. But it is a far richer position. The integrationist does not deny the validity or immense achievements of successful reduction. For some reason (political, social, philosophical?) reductionists seem to need to claim universality for their approach. This book will give them some cause to re-think that position — or so I would hope. Denis Noble CBE FRS hon FRCP Professor of Cardiovascular Physiology at Oxford University Secretary-General, International Union of Physiological Sciences (1993—200Í)

2

Noble, D. (2002), Biological Computation. In: Encyclopedia of Life Sciences, http://www.els.net, London: Nature Publishing Group. Noble, D. (2002), The Rise of Computational Biology. Nature Reviews Molecular Cell Biology, 3, 460-3. See also Novartis Foundation (2002), In Silico Simulation of Biological Processes. Novartis Foundation Symposium, vol. 247, London: John Wiley.

Acknowledgements

Many people have assisted us in writing this book, and we have benefited from their advice, critical comments and constructive suggestions. We are grateful to Dr Hanoch Benyami, Professor Jonathan Dancy, Proiessor John Dupré, D r Hanjo Glock, Professor Oswald Hanfling, Professor Paul Harris, Dr Tim Horder, Professor Andrew Parker, Professor Herman Philipse, Dr John Richardson, Wolfram Schmitt and Professor Stuart Shanker for their helpful comments on one or more (sometimes many more) chapters of the book. W e are especially indebted to Dr John Hyman and Professor Hans Oberdiek, who read and commented constructively on the whole text. We thank Sir Anthony Kenny, Professor Sir Peter Strawson and Professor Georg Henrik von Wright for their encouragement and moral support throughout the writing of this book. We are grateful to Jean van Altena for copy-editing our typescript with her characteristic skill, tact and good humour. Chapters 1 and 2 are a rewritten version of a review article entitled 'The motor system in neuroscience: a history and analysis of conceptual developments', published in Progress in Neurobiology, 67 (2002), pp. 1—52. Parts of chapters 3, 4 and 5 were published in an article entitled 'Perception and memory in neuroscience: a conceptual analysis', in Progress in Neurobiology, 65 (2001), pp. 499—543. Parts of chapter 10 were published in an article by Hacker entitled 'Is there anything it is like to be a bat?' in Philosophy, 77 (2002), pp. 153— 70. We thank the editors of these journals for permission to republish this work. M. R. Bennett, University of Sydney P. M. S. Hacker, St John's College, Oxford

Introduction

Philosophical Foundations of Neuroscience presents the fruits of a cooperative project between a neuroscientist and a philosopher. It is concerned with the conceptual foundations of cognitive neuroscience — foundations constituted by the structural relationships among the psychological concepts involved in investigations into the neural underpinnings of human cognitive, affective and volitional capacities. Investigating logical relations among concepts is a philosophical task. Guiding that investigation down pathways that will illuminate brain research is a neuroscientific one. Hence our joint venture. If we are to understand the neural structures and dynamics that make perception, thought, memory, emotion and intentional behaviour possible, clarity about these concepts and categories is essential. Both authors, coming to this investigation from very different directions, found themselves puzzled by, and sometimes uneasy with, the use of psychological concepts in contemporary neuroscience. The puzzlement was often over what might be meant by a given neuroscientist's claims concerning the brain and the mind, or over why a neuroscientist thought that the experiments he had undertaken illuminated the psychological capacity being studied, or over the conceptual presuppositions of the questions asked. The unease was produced by a suspicion that in some cases concepts were misconstrued, or misapplied, or stretched beyond their defining conditions of application. And the more we probed, the more convinced we became that, despite the impressive advances in cognitive neuroscience, not all was well with the general theorizing. Empirical questions about the nervous system are the province of neuroscience. It is its business to establish matters of fact concerning neural structures and operations. It is the task of cognitive neuroscience to explain the neural conditions that make perceptual, cognitive, cogitative, affective and volitional functions possible. Such explanatory theories are confirmed or infirmed by experimental investigations. By contrast, conceptual questions (concerning, for example, the concepts of mind or memory, thought or imagination), the description of the logical relations between concepts (such as between the concepts of perception and sensation, or the concepts of consciotisness and self-consciousness), and the examination of the structural relationships between distinct conceptual fields (such as between the psychological and the neural, or the mental and the behavioural) are the proper province of philosophy.

2

Introduction

Conceptual questions antecede matters of truth and falsehood. They are questions concerning our forms of representation, not questions concerning the truth or falsehood of empirical statements. These forms are presupposed by true (and false) scientific statements and by correct (and incorrect) scientific theories. They determine not what is empirically true or false, but rather what does and what does not make sense. Hence conceptual questions are not amenable to scientific investigation and experimentation or to scientific theorizing. For the concepts and conceptual relationships in question are presupposed by any such investigations and theorizings. Our concern here is not with trade union demarcation lines, but with distinctions between logically different kinds of intellectual inquiry. (Methodological objections to these distinctions are examined in chapter 14.) Distinguishing conceptual questions from empirical ones is of the first importance. When a conceptual question is confused with a scientific one, it is bound to appear singularly refractory. It seems in such cases as if science should be able to discover the truth of the matter under investigation by theory and experiment — yet it persistently fails to do so. That is not surprising, since conceptual questions are no more amenable to empirical methods of investigation than problems in pure mathematics are solvable by the methods of physics. Furthermore, when empirical problems are addressed without adequate conceptual clarity, misconceived questions are bound to be raised, and misdirected research is likely to ensue. For any unclarity regarding the relevant concepts will be reflected in corresponding unclarity in the questions, and hence in the design of experiments intended to answer them. And any incoherence in the grasp of the relevant conceptual structure is likely to be manifest in incoherences in the interpretation of the results of experiments. Cognitive neuroscience operates across the boundary between two fields, neurophysiology and psychology, the respective concepts of which are categorially dissimilar. The logical or conceptual relations between the physiological and the psychological are problematic. Numerous psychological concepts and categories of concepts are difficult to bring into sharp focus. The relations between the mind and the brain, and between the psychological and the behavioural, are bewildering. Puzzlement concerning these concepts and their articulations, and concerning these apparent 'domains 1 and their relations, has characterized neurophysiology since its inception (we shall begin our investigations in chapter 1 with a historical survey of the early development of neuroscience). In spite of the great advances in neuroscience at the beginning of the twentieth century at the hands of Charles Sherrington, the battery of conceptual questions popularly known as the mind—body or mind—brain problem remained as intractable as ever - as is evident in the flawed Cartesian views embraced by Sherrington and by such of his colleagues and proteges as Edgar Adrian, John Eccles and Wilder Penfield. Brilliant though their work unquestionably was, deep conceptual confusions remained - as we show in chapter 2. Whether the current generation of neuroscientists has successfully overcome the conceptual confusions of earlier generations, or whether it has merely replaced one conceptual entanglement by others, is the subject of our investigation in this book. One such tangle is evident in the persistent ascription of psychological attributes to the brain. For, while Sherrington and his proteges ascribed psychological attributes to the mind (conceived as a peculiar, perhaps immaterial, substance distinct from the brain),

Introduction

3

contemporary neuroscientists tend to ascribe the same range of psychological attributes to the brain (commonly, although not uniformly, conceived to be identical with the mind). But the mind, we argue (§3.10), is neither a substance distinct from the brain nor a substance identical with the brain. And we demonstrate that ascription of psychological attributes to the brain is incoherent (chapter 3). Human beings possess a wide range of psychological powers, which are exercised in the circumstances of life, when we perceive, think and reason, feel emotions, want things, form plans and make decisions. The possession and exercise of such powers define us as the kinds of animals we are. We may enquire into the neural conditions and concomitants for their possession and exercise. This is the task of neuroscience, which is discovering more and more about them. But its discoveries in no way affect the conceptual truth that these powers and their exercise in perception, thought and feeling are attributes of human beings, not of their parts - in particular, not of their brains. A human being is a psychophysical unity, an animal that can perceive, act intentionally, reason and feel emotions, a language-using animal that is not merely conscious, but also self-conscious — not a brain embedded in the skull of a body. Sherrington, Eccles and Penfield conceived of human beings as animals in whom the mind, which they thought of as the bearer of psychological attributes, is in liaison with the brain. It is no advance over that misconception to suppose that the brain is a bearer of psychological attributes. Talk of the brain's perceiving, thinking, guessing or believing, or of one hemisphere of the brain's knowing things of which the other hemisphere is ignorant, is widespread among contemporary neuroscientists. This is sometimes defended as being no more than a trivial fagon de parley. But that is quite mistaken. For the characteristic form of explanation in contemporary cognitive neuroscience consists in ascribing psychological attributes to the brain and its parts in order to explain die possession of psychological attributes and the exercise (and deficiencies in die exercise) of cognitive powers by human beings. The ascription of psychological — in particular, cognitive and cogitative — attributes to the brain is, we show, also a source of much further confusion. Neuroscience can investigate the neurai conditions and concomitants of the acquisition, possession and exercise of sentient powers by animals. It can discover the neural preconditions for the possibility of the exercise of distinctively human powers of thought and reasoning, of articulate memory and imagination, of emotion and volition. This it can do by patient inductive correlation between neural phenomena and the possession and exercise of psychological powers, and between neural damage and deficiencies in normal mental functions. What it cannot do is replace the wide range of ordinary psychological explanations of human activities in terms of reasons, intentions, purposes, goals, values, rules and conventions by neurological explanations (reductionism is discussed in chapter 13). And it cannot explain how an animal perceives or thinks by reference to the brain's, or some part of the brain's, perceiving or thinking. For it makes no sense to ascribe such psychological attributes to anything less than the animal as a whole. It is the animal that perceives, not parts of its brain, and it is human beings who think and reason, not their brains. The brain and its activities make it possible for us - not for it - to perceive and think, to feel emotions, and to form and pursue projects. While the initial response of many neuroscientists to the accusation of conceptual confusion is to claim that the ascription of psychological predicates to the brain is a mere

4

Introduction

facón de parler, their reaction to the demonstrable fact that their explanatory theories nontrivially ascribe psychological powers to the brain is sometimes to suggest that this error is unavoidable due to the deficiencies of language. We confront this misconception in chapter 14, where we show that the great discoveries of neuroscience do not require this misconceived form of explanation - that what has been discovered can readily be described and explained in our existing language. W e demonstrate this by reference to the much discussed phenomena resultant upon commissurotomy, described (or, we suggest, misdescribed) by Sperry, Gazzaniga and others (§14.3). In Part II we investigate the use of concepts of perception, memory, mental imagery, emotion and volition in current neuroscientific theorizing. From case to case we show that conceptual unclarity - failure to give adequate attention to the relevant conceptual structures — has often been die source of theoretical error and the grounds for misguided inferences. It is an error, a conceptual error, to suppose that perception is a matter of apprehending an image in the mind (Crick, Damasio, Edelman), or the production of a hypothesis (Helmholtz, Gregory), or the generation of a 3-D model description (Marr). It is confused — a conceptual confusion - to formulate the binding problem as the problem of combining data of shape, colour and motion to form the image of the object perceived (Crick, Randel, Wurtz). It is wrong, conceptually wrong, to suppose that memory is always of the past, or to think that memories can be stored in the brain in the form of the strength of synaptic connections (Kandel, Squire, Bennett). And it is mistaken, conceptually mistaken, to suppose that the investigation of thirst, hunger and lust is an investigation into the emotions (Rolls) or to think that the function of the emotions is to inform us of our visceral and musculoskeletal state (Damasio). The initial reaction to such critical remarks may well be indignation and incredulity. H o w can a flourishing science be fundamentally in error? H o w could there be unavoidable conceptual confusion in a well-established science? Surely, if there are problematic concepts, they can easily be replaced by others that are unproblematic and that serve the same explanatory purposes. Such responses betoken a poor understanding of the relation between form of representation and facts represented, and a misunderstanding of the nature of conceptual error. They also betray ignorance of the history of science in general, and of neuroscience in particular. Science is no more immune to conceptual error and confusion than any other form of intellectual endeavour. The history of science is littered with the debris of theories that were not simply factually mistaken, but conceptually awry. Stahl's theory of combustion, for example, was conceptually flawed in ascribing, in certain circumstances, negative weight to phlogiston - an idea that made no sense within its framework of Newtonian physics. Einstein's famous criticisms of the theory of electromagnetic aether (the alleged medium by which light was thought to be propagated) were directed not only at the results of the Michelson-JVtorley experiment, which had failed to detect any effect of absolute motion, but also at a conceptual confusion concerning relative motion involved in the role ascribed to aether in the explanation of electromagnetic induction. Neuroscience has been no exception — as we show in our historical survey. It is true enough that the subject is now a flourishing science. But that does not render it immune to conceptual confusions and entanglements. Newtonian kinematics was a flourishing science, but that did not stop

Introduction

5

]Sfewton from becoming entangled in conceptual confusions over the intelligibility of action at a distance, or from bafflement (not remedied until Hertz) over the nature of force. So too, Sherrington's towering achievement in explaining the integrative action of synapses in the spinal cord, and thereby eliminating, once and for all, the confused idea of a 'spinal soul', was perfectly compatible with conceptual confusions concerning the 'cerebral soul' or mind and its relation to the brain. Similarly, Penfield's extraordinary achievements in identifying functional localization in the cortex, as well as in developing brilliant neurosurgical techniques, were perfectly compatible with extensive confusions about the relation between the mind and the brain and about the 'highest brain function' (an idea borrowed from Hughlings Jackson). In short, conceptual entanglement can coexist with nourishing science. This may appear puzzling. If the science can flourish despite such conceptual confusions, why should scientists care about them? Hidden reefs do not imply that the seas are not navigable, only that they are dangerous. The moot question is how running on these reefs is manifest. Conceptual confusions may be exhibited in different ways and- at different points in the investigation. In some cases, the conceptual unclarity may affect neither the cogency of the questions nor the fruitfulness of the experiments, but only the understanding of the results of the experiments and their theoretical implications. So, for example, Newton embarked on the Optics in quest of insight into the character of colour. The research was a permanent contribution to science. But his conclusion that 'colours are sensations in the sensorium' demonstrates failure to achieve the kind of understanding he craved. For, whatever colours are, they are not 'sensations in the sensorium'. So in so far as Newton cared about understanding the results of his research, then he had good reason for caring about the conceptual confusions under which he laboured — for they stood in the way of an adequate understanding. In other cases, however, the conceptual confusion does not so happily bracket the empirical research. Misguided questions may well render research futile (examples will be examined in relation to mental imagery (§6.3.1) and voluntary movement (§8.2)). Rather differently, misconstrual of concepts and conceptual structures will sometimes produce research that is by no means futile, but that fails to show what it was designed to show (examples will be discussed in relation to memory (§§5.2.1—5.2.2) and to emotions and appetites (§7.1)). In such cases, the science may not be flourishing quite as much as it appears to be. It requires conceptual investigation to locate the problems and to eliminate them. Are these conceptual confusions unavoidable'? Not at all. The whole point of writing this book is to show how to avoid them. But, of course, they cannot be avoided while leaving eveiything else intact. They can be avoided - but if they are, then certain kinds of questions will no longer be asked, since they will be recognized as resting on a misunderstanding. As Hertz put it in the wonderful introduction to his Principles of Mechanics: 'When these painful contradictions are removed, . . . our minds, no longer vexed, will cease to ask illegitimate questions.' Equally, certain kinds of inferences will no longer be drawn from a given body of empirical research, since it will be realized to have little or no bearing on the matter which it was nreant to illuminate, even though it may bear on something else.

6

Introduction

If there are problematic concepts, can they not be replaced by others that serve the same explanatory function? A scientist is always free to introduce new concepts if he finds existing ones inadequate or insufficiently refined. But our concern in this book is not with the use of new technical concepts. We are concerned with the misuse of old, nontechnical concepts — concepts of mind and body, thought and imagination, sensation and perception, knowledge and memory, voluntary movement, and consciousness and selfconsciousness. There is nothing inadequate about these concepts relative to the purposes they serve. There is no reason for thinking that they need to be replaced in the contexts that are of concern to us. What are problematic are neuroscientists' misconstruals of them and the misunderstandings consequently engendered. These are remediable by a correct account of the logico-grammatical character of the concepts in question. And this is what we have tried to supply. Granted that neuroscientists may not be using these common or garden concepts the way the man in the street does, with what right can philosophy claim to correct them? How can philosophy so confidently judge the clarity and coherence of concepts as deployed by competent scientists? H o w can philosophy be in a position to claim that certain assertions made by sophisticated neuroscientists make no sense? We shall resolve such methodological qualms in the following pages. But some initial clarification here may remove some doubts. What truth and falsity is to science, sense and nonsense is to philosophy. Observational and theoretical error result in falsehood; conceptual error results in lack of sense. How can one investigate the bounds of sense? Only by examining the use of words. Nonsense is often generated when an expression is used contrary to the rules for its use. The expression in question may be an ordinary, non-technical expression, in which case the rules for its use can be elicited from its standard employment and received explanations of its meaning. Or it may be a technical term of art, in which case the rules for its use must be elicited from the theorist's introduction of the term and the explanations he offers of its stipulated use. Both kinds of term can be misused, and when they are, nonsense ensues — a form of words that is excluded from the language. For either nothing has been stipulated as to what the term means in the aberrant context in question, or this form of words is actually excluded by a rule specifying that there is no such thing as . . . (e.g. that there is no such thing as 'east of the North Pole'), that this is a form of words that has no use. Nonsense is also commonly generated when an existing expression is given a new, perhaps technical or quasi-technical use, and the new use is inadvertently crossed with the old — for example, inferences are drawn from propositions containing the new term which could only licitly be drawn from the use of the old one. It is the task of the conceptual critic to identify such transgressions of the bounds of sense. It is, of course, not enough to show that a certain scientist has used a term contrary to its ordinary use — for he may well be using the term in a new sense. The critic must show that the scientist intends using the term in its customary sense and has not done so, or that he intends using it in a new sense but has inadvertently crossed the new sense with the old. The wayward scientist should, whenever-possible, be condemned out of his own mouth. W e address methodological qualms in detail both in chapter 3, section 3, and in chapter 14. The final misconception against which we wish to warn is the idea that our reflections are unremittingly negative. All we are concerned with, it might be thought, is criticizing.

Introduction

7

Our work may appear at a superficial glance to be no more than a destructive undertaking that promises neither assistance nor a hew way forward. Worse, it may even appear to be engineering a confrontation between philosophy and cognitive neuroscience. Nothing could be further from the truth. We have written this book in admiration for the achievements of twentieth-century neuroscience, and out of a desire to assist the subject. But the only ways in which a conceptual investigation can assist an empirical subject are by identifying conceptual error (if it obtains) and by providing a map that will help prevent empirical researchers from wandering off the high roads of sense. Each of our investigations has two aspects to it. On the one hand, we have tried to identify conceptual problems and entanglements in important current theories of perception, memory, imagination, emotion and volition. Moreover, we argue that much contemporary writing on die nature of consciousness and self-consciousness is bedevilled by conceptual difficulties. This aspect of our investigations is indeed negative and critical. O n the other hand, we have endeavoured, from case to case, to provide a perspicuous representation of the conceptual field of each of the problematic concepts. This is a constructive endeavour. We hope that these conceptual overviews will assist neuroscientists in their reflections antecedent to the design of their experiments. However, it cannot be the task of a conceptual investigation to propose empirical hypotheses that might solve the empirical problems faced by scientists. To complain that a philosophical investigation into cognitive neuroscience has not contributed a new neuroscientific theory is like complaining to a mathematician that a new theorem he has proved is not a new physical theoiy. It is improbable that many neuroscientists will wish to read a 450-page conceptual investigation from cover to cover. Consequendy, we have tried to make our chapters on select psychological concepts as self-contained as possible. We have done this in the hope that the book will serve as a conceptual reference work for cognitive neuroscientists who wish to check the contour lines of the psychological concept relevant to their investigation. This has, of course, meant that there is a degree of repetition between certain chapters. This is, we hope, warranted by the objective. The chapters of the book are accompanied by italicized marginalia indicating the subject under discussion in the correlated paragraph or paragraphs. The purpose of this is to facilitate surveyability, to make it easier to follow the steps in the argument, and to assist in locating arguments. The section headings in the table of contents are accompanied by the italicized names of neuroscientists (and occasionally philosophers who concern themselves with neuroscientific and cognitive scientific matters) whose theories are either discussed in some detail or mentioned en passant, in the course of the chapter. This will, we hope, help the reader to locate the themes and discussions that are of specific interest with ease.

I Philosophical Problems in Neuroscience: Their Historical and Conceptual Roots

1

The Early Growth of Neuroscientijic Knowledge: The Integrative Action of the Nervous System

The conceptual framework for early investigations into the biological basis for human sensory, volitional and intellectual capacities was set by Aristotle's philosophical writings on the psuche. The early growth of neuroscientific knowledge was dominated by the question of how the contraction of muscles involved in voluntary movements of limbs is effected. However, Aristotle's own rudimentary investigations, which led him to believe that the blood vessels initiate muscle contraction, were a false start. It was, above all, Galen's much later discoveries of the nerve supply to muscles from the spinal cord that made it clear that it is the nerves that carry out this function. Galen's work initiated 2,000 years of enquiry into how the spinal cord and brain are involved in voluntary movement and into the reflex origins of some movements. The identification of motor and sensory spina) nerves, the role of the spinal cord in reflex movements, and the relationship between the action of the brain and the spinal cord in voluntary and reflex movement were all resolved by experiments. These involved observations on muscles and limbs following lesions to different parts of the nervous system. In this way a conception evolved of how the functions of the brain, spinal cord and nerves are integrated to give the final motor output. The conceptual framework within which neuroscientific knowledge grew originated in Aristotelian thought, but it was subsequently transformed by the Cartesian revolution in the seventeenth century. In this chapter we shall adumbrate the development of ideas concerning the neural basis of animate functions, concentrating increasingly upon what Sherrington, the greatest of neuroscientists, called 'the integrative action of the nervous system', as it applies to movement. This sketch of the history of the slow growth of knowledge about the nervous system and its operations will display some of the conceptual difficulties encountered by natural philosophers over the centuries as they grappled with the problems concerning the biological foundations of characteristic powers of animate beings in general, and humans in particular. As we shall see, the roots of current

Philosophical Problems in Neuroscience

12

conceptual difficulties in cognitive neuroscience are buried deep in the past. Grasping this aspect of our intellectual and scientific heritage will help to bring current conceptual problems into sharp focus. These problems are the principal concern of this book. It might be asked why we do not concentrate more on the role of the great sensory systems, such as vision, in our historical sketch of the integrative action of the nervous system. The reason is that the early neuroscientists took up the challenge of understanding the motor system first, for it aDowed experimentation which they could undertake to test their ideas with the techniques then available. This was not the case with the sensory systems. These pioneers saw the need to integrate their account of the sensory systems into their evolving knowledge of muscular contraction and movement. This led them, to speculate on the relationship between vision and motor performance. It did not, however, add much to our understanding of how vision occurs, a subject that had to wait for techniques that became available only in the nineteenth and especially the twentieth centuries.

1.1 Aristotle, Galen and Nemesíus: The Origins of the Ventricular Doctrine Aristotle's conception of the psuche Aristotle is the first great biologist whose treatises and observational data survive. His philosophical world picture shaped European thought until and, in certain respects, beyond the scientific revolution of the seventeenth century. So although his knowledge of the nervous system was almost non-existent, his fundamental conceptions of animate life are indispensable to an understanding of the reasoning of the early scientists, such as Galen and Nemesius, who probed the nature of the nervous system and its role in determining the cognitive, cogitative, affective and volitional powers of man. Moreover, as we shall see, Aristotle's conception of the nature of man, of the relation between organs and functions, between the body and the distinctive capacities that constitute what he called 'the psuche' was profound. The Aristotelian conception of the psuché and the Cartesian conception of the mind, which displaced it in the seventeenth century, constitute in certain respects two fundamentally different ways of thinking about human nature, which have informed neuroscientific reflection on the integrative action of the nervous system throughout the ages. „ , , r r , I he psuche as the form of the , , , natural body

Aristotle ascribed to each living organism a psuche. The ,_ , . . . . , , , psuche was conceived to be the form of a natural body r J J . * that has life. It was also characterized as the Jirst actuality of a natural body that has organs (DA 412b5—6). Aristotle's technical terminology needs elucidation. In its common meaning, 'psuche' signified 'breath' or 'life breath' (which one 'expires' at the moment of death or in a faint), as did the later Latin term 'anima', by which it was translated. It is linked with the idea of wind and of vital power. It was a pre-Aristotelian philosophical innovation to detach psuché from such associations. It was an Aristotelian 1

Aristotle, De Anima 412520. Subsequent references to this treatise in the text will be flagged 'DA'.

The Early Growth of Neuvoscientijic Knowledge

13

innovation to link it firmly to all organisms as the principle of life that informs each living being. Although 'psuche' is commonly translated as 'soul', it is important to realize that, as used by Aristotle, 'psuche' has none of the religious and ethical connotations of our term 'soul'. Psuche is 'the principle of animal life' (DA 402 3 7-8), and indeed of vegetal life too. For plants, no less than animals, have a psuche. It would be equally misleading to translate 'psuche' as 'mind', since the mind and mental powers are not associated, as psuche is, with growth, nutrition or reproduction, which characterize all forms of living things. Nor is psuche essentially linked with consciousness, as is the Cartesian conception of the mind. The term 'psuche', which, in conformity with tradition, we shall in the sequel translate as 'soul', is a biological concept — not a religious or an ethical one. It will be important to keep this in mind not only in regard to Aristotle, but also in respect of neuroscientific debates in the seventeenth and eighteenth centuries on the existence of a 'spinal soul' (see §1.4). Aristotle introduced the distinction between form and Form distinguished from matter . . , c * matter to provide the theoretical apparatus necessary lor describing persistence through change. Substances, irrespective of whether they are spaceoccupying things of a given kind (such as a rock, a tree, a horse, an axe or a man) or whether they are partitioned quantities of stuff of a certain kind (of water, bronze, wine or cheese, which may occur as a drop, a nugget, a bottle or a slice) undergo change. The change may be 'accidental' or 'substantial' {'essential'). 'Accidental change' is a change to the non-essential properties ('accidents') of a given substance. It is the acquisition of a new attribute or the loss of an existing attribute — as when a tree grows taller, an axe becomes blunter, a man becomes fatter, and as when a pool of water becomes warm, or a certain quantity of gold melts. Accidental change is compatible with the continued identity of the substance. Substantial change occurs when a substance changes its essential properties, as when wine turns to vinegar, or milk to cheese, and equally, when a living creature dies. Substantial change is incompatible with the continued existence of the substance in question — if the wine changes into vinegar, then there is no more wine, and if a horse dies, it ceases to exist, and only its remains, a dead body, are left behind. Aristotle introduced the notion of matter as a technical term to pick out that which has a capacity for substantial change, and form to pick out that which makes a certain matter into the kind of substance it is. Accordingly there are both accidental forms (e.g. the accidental forms of the different colours which a thing may acquire or lose, while remaining the very same thing), and substantial forms (e.g. of wine or vinegar, of a plant or of a man). When a substance undergoes accidental change, it retains its substantial form through change. _ , It is important to realize that although individual things torm and matter are not parts of . ,. , . , , , , . , r ,. or any given kind are said to be both torm and matter, form and matter are not parts of a thing. Matter cannot exist without form — its form may change, accidentally or essentially, but it must have some form or other. Equally, form cannot exist without matter — for the form of X-ness to exist is just for there to be some substance that is X. It may well be argued that this conceptual apparatus is well suited to discussing stuffs and their transformation (e.g. milk to cheese), and perhaps also (with adjustment) to investigating things and their constitutive stuffs (e.g. a sword and the steel from which it is made), as well as (with different

14

Philosophical Problems in Neuroscience

adjustments) things and their constitutive parts (e.g. a h o u s e a n d the bricks of w h i c h it is m a d e ) . It is questionable w h e t h e r it is very well suited to b e i n g e x t e n d e d to the descript i o n of things and their capacities. 2 Be that as it may, Aristotle did so e x t e n d it i n characterizing the soul as the form of the living b o d y , and in simultaneously conceiving of t h e soul as the first actuality of t h e living b o d y . . The psuche of a creature is consti, , . r ,• • tuted by its jirst actualities

T h e actualities (entelechiat) of a substance are those things , . , . . . , . . , w h i c h it is or is d o i n g at a given t i m e . A m o n g its actu, alities are (rather confusingly) its p o w e r s (i.e. the active a n d passive potentialities w h i c h it actually possesses). Such p o w e r s m a y b e possessed w i t h o u t being exercised at a given time - t h e sighted d o n o t b e c o m e blind w h e n they sleep, and English speakers d o n o t cease to k n o w English w h e n they are silent. Aristotle refers to the unexercised dispositional p o w e r (hexis) of an animal as a first actuality (e.g. b e i n g sighted, k n o w i n g English), b y contrast w i t h t h e exercise o f a dispositional p o w e i (the actual activity of, for example, w a t c h i n g s o m e t h i n g or speaking English), w h i c h h e refers to as a second actuality (energeia). "When the soul is said to b e the actuality of the b o d y , it is so only as first actuality. For a h v i n g creature has a soul — that is, possesses its distinctive defining p o w e r s — w h e n it is asleep n o less than w h e n it is a w a k e . So, t h e phrase 'to have a soul' does n o t signify a relation . ., , . , , r of rpossession b e t w e e n an agent and an entity, as does to J , ° have a car . F u r t h e r m o r e , the soul does n o t stand to the ' b o d y as the brain does, for it is n o t a part of the b o d y . H o w , t h e n , should these things b e conceived? Aristotle gives a pair of analogies. T h e matter of an axe is the w o o d a n d i r o n of w h i c h it is m a d e . Its form is its capacity to c h o p . T h e axe's first actuality is its p o w e r to c h o p w o o d , w h i c h it possesses inasmuch as its constituent matter has b e e n appropriately fashioned into blade a n d handle. Its p o w e r to c h o p c a n n o t exist i n d e p e n d e n d y o f t h e m a t t e r ( w o o d a n d iron) o r t h e parts (handle a n d blade) of w h i c h it consists. (But, of course, an axe is an inanimate artefact, and inanimate things have n o soul.) Similarly, Aristotle suggests, w e can c o m p a r e the relationship b e t w e e n an animal and its soul w i t h the relationship b e t w e e n an eye and the p o w e r of sight. If the eye w e r e an animal, as it w e r e , then its soul w o u l d be sight (DA 412 b 18); although, of course, the eye is n o t an animal, b u t a part of an animal - a n d accordingly, w h i l e it has a function, it has n o soul. . T h e soul consists of the essential, defining functions of a The psuche or soul, as first actual,. . , . ., 1^1 -1 r s r e v i v a i 0 f a misconceived idea of the great nineteenth-century mathematical logician Gottlob Frege.14 Frege distinguished be-, tween the perceptible 'outer world' of physical objects, the private 'inner world' of mental entities, and a 'third realm' of thoughts (propositions) that are imperceptible by the senses, but nevertheless public and shareable. Popper followed suit, distinguishing between World 1 of physical things, World 2 of mental things, and World 3 of thoughts, theorems, theories and other abstracta. The conception is confused, since although we distinguish material objects from mental states, and both from propositions or theorems, these do not collectively constitute 'worlds' in any sense whatsoever. Furthermore, neither mental states1

12

J. C. Eccles, in K. R. Popper and J. C. Eccles, The Self and its Brain (Springer Verlag, Berlin, 1977), p. 357. 13 Eccles, Human Mystery, p. 3. Subsequent references in che text to this book will be flagged 'HM. 14 Gottlob Frege, 'The thought', in his Collected Papers on Mathematics, Logic and Philosophy (Bh Oxford, 1984), pp. 351-72.

Sherrington and his Proteges

51

propositions are denizens of a distinct 'world'. There is only one world, which is described by specifying whatever is (contingently) the case. We do indeed talk of people's mental states of cheerfulness or depression, or of their having toothache. But this does not . « ¡ y that cheerfulness, depression or toothache are peculiar mental entities that exist in 'inner world'. These nomináis ('cheerfulness', 'depression', 'toothache') merely provide an indirect way of talking of people being cheerful or depressed and of their tooth's hurting - it introduces no new entities, merely new ways of talking about existing entities (e.g. about people and how things are with them). Similarly, we talk of propositions, theorems and other abstracta - but this too only appears to introduce new entities, and is really no more than a convenient way of talking about what is or might be said, asserted, or proved, etc There is absolutely no need to succumb to Platonism and conjure new entities into existence and new worlds for them to inhabit. All talk of expressions standing for 'abstract entities' is a misleading way of saying that expressions that look as if they stand for concrete entities do not do so at all, but rather fulfil quite different functions. To be sure, this does not mean that there are n o mental states, no cheerfulness, depression or anxiety, or that there are no propositions, no theories or theorems. O n the contrary, it means that there are - only they are not kinds of entities. Popper's three-world doctrine impressed Eccles, and he formulated his dualism in terms of it. World 1, the material world of the cosmos, he declared, consists of mere material things and of beings that enjoy mental states. The latter, being a subset of the entities in World 1, he refers to collectively as 'World 1M'. This 'world' stands in reciprocal causal interaction with World 2 by means of what he terms 'the liaison brain' (HM 211). „, _, , rrjr , , , Research done by Kornhuber and his colleagues on The ¡npatt on hales of Komhuber s , . . . . . , , , , ,. . ,- j changes in electrical potential antecedent to a voluntary r research on leaainess potential ' movement had revealed that the so-called readiness potential began up to 800 milliseconds before the onset of the muscle action potential, and led to a sharper potential, the pre-motion positivity, beginning at 80-90 milliseconds prior to the movement. The patterns of neuronal discharges eventually project to the appropriate pyramidal cells of the motor cortex and synaptically excite them to discharge, so generating the motor potential (a localized negative wave) just preceding the motor pyramidal cell discharge that initiates the movement. The question on which Kornhuber's research seemed to throw light was: 'How can willing of a muscular movement set in train neuronal events that lead to the discharge of pyramidal cells of the motor cortex and so to the activation of the neuronal pathways that lead to the muscle contraction?' (HM 214). It is striking that Eccles took these discoveries to betoken empirical confirmation of mind-brain interaction of a kind (but in a different location) that had been envisaged by Descartes. He argued as follows: What is happening in my brain at a time when the willed action is in the process of being carried out? It can be presumed that during the readiness potential there is a developing specificity of the patterned impulse discharges in neurons so that eventually there are activated the pyramidal cells in the correct motor cortical areas for bringing about the required movement. The readiness potential can be regarded as the neuronal counterpart of the voluntary intention. The surprising feature of the readiness potential is

52

Philosophical Problems in Neuroscience

its very wide extent and gradual build u p . Apparently, at the stage of willing a movement, there is a very wide influence of the self-conscious mind on the pattern of module operation. Eventually this immense neuronal activity is moulded and directed so that it concentrates onto the pyramidal cells in the proper zones of the motor cortex for carrying out the required movement. T h e duration of the readiness potential indicates that the sequential activity of the large numbers of modules is involved in the long incubation time required for the self-conscious mind to evoke discharges from the motor pyramidal cells. . . . It is a sign that the action of the self-conscious mind o n the brain is not of demanding strength. W e may regard it as being more tentative and subtle, and as requiring time to build up patterns of activity that may be modified as they develop (HAf 217) So, Eccles c o n c e i v e d of w h a t h e called 'the duahs . . , , . , , . . , , ( interactiomst hypothesis as h e l p i n g to resolve and redeii ,• , , 1 r line the p r o b l e m or a c c o u n t i n g tor the long duration of the readiness potential that precedes a voluntary action' (HM 217). Descartes, as w e have noted, conceived of the m i n d as operating u p o n the pineal gland to l generate the m i n u t e fluctuations in the animal spirits (the equivalent of neural transmitters) ,J in the ventricle in w h i c h h e t h o u g h t the pineal gland was suspended. This, h e held, enabled t h e acts of will of the m i n d to affect the m o t i o n s of the animal spirits, w h i c h are " t h e n transmitted to the muscles. B u t the question o f how an immaterial substance could actually interact causally w i t h a material object such as t h e pineal gland to p r o d u c e the appropriate m i n u t e m o t i o n s was left totally u n a n s w e r e d . In m u c h the same way, Eccles t h o u g h t that the 'self-conscious m i n d ' interacts causally w i t h the pyramidal cells of the m o t o r cortex, gradually {rather than instantaneously) getting t h e m to discharge. B u t the question of how an immaterial entity such as the m i n d can interact causally w i t h neurons was left equally unanswered. B o t h thinkers erred in conceiving of the m i n d as an •" * entity of s o m e kind. H a d they h e e d e d Aristotle in thinking of the m i n d n o t as an entity b u t as an array of p o w e r s or potentialities, they would have b e e n m u c h closer to the truth, a n d w o u l d n o t have b e c o m e e n m e s h e d in insoluble p r o b l e m s of interaction. For it patently makes n o sense to ask h o w one's abilities to d o the various things o n e can d o interact w i t h one's brain. , , , , B o t h thinkers erred in imagining that voluntary m o v e 3. Misconceptions about the will , , , , , r ments are m o v e m e n t s p r o d u c e d or caused by antecedent acts o f will. 1 5 F o r a l t h o u g h there are such things as acts o f will - namely, acts performed w i t h great effort to o v e r c o m e one's reluctance, aversion or difficulties in acting in adverse circumstances — obviously the vast majority of o u r ordinary voluntary actions involve no 'act of will' in this sense at all. W e shall e x a m i n e this c o n c e p t i o n in chapter 8. Eccles was further confused over the object o f the alleged act of will, w h i c h is variously characterized as (i) a muscular m o v e m e n t , (ii) an action or (iii) a m o v e m e n t of a limb. Cartesian problems recapitulated: . r • 1. interaction

15 This error is still common among neuroscientists, and informs the research of Benjamin Libet and his colleagues that we discuss below (see §§8.1-8.2).

Sherrington and his Proteges

53

It is, of course, possible to i n t e n d to m o v e - for example, ^ ^ ^ ^ ^ ^ ^ .$ S Q m e t h i n g w e rather rarely i n t e n d to d o , and although the m o v e m e n t of muscles is 'nvolved in all o u r positive, physical acts (by contrast w i t h acts of omission and m e n t a l ' \ ^vhat w e intend, a n d w h a t w e voluntarily perform, are actions (such as raising o u r

:Conjusionsabo^ the object of the alleged act of mil

arm w r i o n g a letter, saying s o m e t h i n g , p i c k i n g up a b o o k , reading a b o o k , a n d so on), and not the constitutive muscle m o v e m e n t s of these actions, of w h i c h w e are largely unaware. B u t it is easy to see w h y a neuroscientist w h o is attracted to dualism should confuse the objects of the will. For, according to the dualist conception, the m i n d has causally to affect the brain, a n d the causal p o w e r s of neural events in the brain causally affect muscle contraction. , . Prnhlems of volitional interaction twu'tj between mind and brain

T h i s raises yet a further insoluble p r o b l e m for the dualist. . . . * . ^ n e s e i f _ c o n s C i o u s m i n d is supposed to influence t h e j „ , ,• , pattern of m o d u l e operation, gradually m o u l d i n g a n d directing it so that it concentrates o n the pyramidal cells i n the p r o p e r zones of the m o t o r cortex for carrying o u t the i n t e n d e d m o v e m e n t . B u t h o w does the 'self-conscious m i n d ' know w h i c h pyramidal cells to concentrate on, a n d h o w does it select the p r o p e r zones of the motor cortex? For it w o u l d n e e d such k n o w l e d g e in order to execute such actions. And it is certainly n o t k n o w l e d g e of w h i c h the self-conscious m i n d is conscious. T o these questions there can b e n o answers, any m o r e than the n i n e t e e n t h - c e n t u r y innervationist ideo-motor theories of voluntary m o v e m e n t , favoured b y such e m i n e n t scientists as Helmholtz and M a c h (and psychologists such as Bain a n d W u n d t ) , could answer the question of h o w the m i n d , i n addition to having images of kinaesthetic sensations that allegedly accompany voluntary m o v e m e n t s , directs the currents of energy going from the brain to the appropriate muscles. (There m u s t b e appropriate feelings of i n n e r v a t i o n - of 'impulse' or 'volitional energy', they t h o u g h t , otherwise t h e m i n d could n e v e r tell w h i c h particular current of energy, w h e t h e r the current to this muscle or the current to that o n e , was the right o n e to use.) A second piece of empirical research encouraged Eccles in Ecdes's conception of the implica. . , . . , , . „ , ,. r. ,,, , ,. . , his advocacy of mteractionist dualism, bperry s discoveries r ! twns of Sperry s discoveries about ' results of split-brain operations c o n c e r n i n g the capacities .of spht-bram patients w e r e sinking. H e himself t o o k t h e m to vindicate s o m e form of mind-brain interactionism: Conscious phenomena in this scheme are conceived to interact with and to largely govern the physiochemical and physiological aspects of the brain process. It obviously works the other way round as well, and thus a mutual interaction is conceived between the physiological and the mental properties. Even so, the present interpretation would tend to restore the m i n d to its old prestigious position over matter, in the sense that the mental p h e n o m e n a are seen to transcend the phenomena of physiology and biochemistry. 16

Quoted by Eccles, without a reference, in Popper and Eccles, Self and Us Brain, p. 374.

54

Philosophical Problems in Neuroscience

It is therefore unsurprising that Eccles t h o u g h t that Sperry's w o r k had dramatic implications 'It is m y thesis', h e w r o t e , 'that the philosophical p r o b l e m of brain a n d m i n d has b^eti transformed by these investigations of the functions of t h e separate d o m i n a n t and minor hemispheres in the split-brain subjects' (HM 222). T h e ' m o s t remarkable discovery', Etc les held, was that all the neural activities in t h e right h e m i s p h e r e 'are u n k n o w n to die speaking subject, w h o is only in liaison w i t h the n e u r o n a l activities in the left [dominant] h e m i s p h e r e ' . T o be sure, the right h e m i s p h e r e is 'a very highly developed brain', but it 'cannot express itself in language, so is n o t able to disclose any experience of consciousness that w e can recognize'. T h e d o m i n a n c e of t h e left h e m i s p h e r e , h e argued, is d u e to its verbal a n d ideational abilities, and 'its liaison to self-consciousness ( W o r l d 2 ) ' (HM 220). For w h a t Sperry's w o r k shows, Eccles averred, is 'that only a specialized z o n e ot the cerebral hemispheres is in liaison w i t h the self-conscious m i n d . T h e t e r m liaison brain denotes all those areas of the cerebral cortex that potentially are capable of b e i n g in direct liaison w i t h the self-conscious mind.' 1 7 , Eccles s conception Jof the liaison , . , _ , . r brain and Descartes s conception of the pineal gland compared

Descartes t h o u g h t that the pineal gland was the point of • j j . i • i i i c , contact ot the m i n d and the brain, and that the niiiiti , , , . , . , ^ , 1 1 apprehends w h a t is before t h e eyes of the b o d y m ^ irtuc of the images that c o m e from the t w o eyes and arc u n i t e d o n the pineal gland. Eccles t h o u g h t that t h e liaison brain was t h e p o i n t of contact with the m i n d , w h e r e the n e r v e impulses from the sense-organs are, in some sense, made available to the m i n d . B u t there is an interesting difference b e t w e e n the t w o doctrines. Descartes t h o u g h t that the pineal gland itself— that is, a part of the brain — fulfils the task of the Aristotelian and scholastic sensus communis, the task of synthesizing and unifying the data o f the separate senses. I n this respect, his t h o u g h t was m o r e up to date than Ecdes's, since c o n t e m p o r a r y neuroscientists think likewise that the ' b i n d i n g p r o b l e m ' is solved by the brain (rather than by the mind). 1 8 For Singer's discoveries 1 9 of c o h e r e n t oscillatory firings in disparate parts of the brain c o n c o m i t a n t w i t h perceptual experience suggesr that the simultaneity of these manifold neuronal activities a n d their connections to o t h e r are.i«. of the cortex are necessary conditions for a perceiver to have the kind of unified perceptual experience w e have. Eccles, b y contrast, defended w h a t h e called ' t h e strong dualist hypothesis' that the self-conscious mind is actively engaged in reading out from the multitude of active modules at the highest levels of the brain, namely in the liaison areas that are largely in the dominant cerebral hemisphere. T h e self-conscious mind selects from these modules according to attention, and from m o m e n t to m o m e n t integrates its selection to give

17

Ibid., p. 358. See, e.g., F. Crick, The Astonishing Hypothesis (Touchstone, London, 1995), pp. 22, 232 .uid E. Kandel and R . Wurtz, 'Constructing the visual image', in Kandel, Schwartz and Jessell iixK). Principles of Neural Science (Elsevier, New York, 2001), pp. 492, 502. (The binding problem is discussed in §4.2,3 below.) Iy A. K. Engel, P. R. Roelfsema, P. Fries, M. Brecht and W . Singer, 'Role of the temporal donuiti for response selection and perceptual binding', Cerebral Cortex, 6 (1997), pp. 571-82. 18

Sherrington and his Proteges

55

initv even to the most transient experience. Furthermore, the self-conscious m i n d acts on these modules, modifying the dynamic spatio-temporai patterns of the neuronal events. Thus the self-conscious mind exercises a superior interpretative and controlling role upon the neuronal events b o t h within the modules and between the modules. /} key component of the hypothesis is that the unity of conscious experience is provided by the self-conscious mind and not by the neuronal machinery of the liaison areas of the cerebral cortex. Hitherto it has been impossible to develop any neurophysiological theory that explains how a diversity of brain events comes to be synthesized so that there is a unified conscious experience . . . M y general hypothesis regards the neuronal machinery as a multiplex of radiating and receiving structures (modules). T h e experienced unity comes, not from a neurophysiological synthesis, b u t from the proposed integrating character of the self-conscious mind. I conjecture that in the first place the raison d'etre of the selfconscious mind is to give this unity of the self in all its conscious experiences and actions. (BM 227f.) H o w does the m i n d engage in this activity o f synthesis (or 'binding')? Eccles suggested that the m i n d plays through the whole liaison brain in a selective and unifying manner. T h e analogy is provided by a searchlight. Perhaps a better analogy would be some multiple scanning and probing device that reads o u t from and selects from the immense and diverse patterns of activity in the cerebral cortex and integrates these selected components, so organizing them into the unity of conscious experience. . . . Thus I conjecture that the selfconscious mind is scanning the modular activities in the liaison areas of the cerebral cortex. . . . From m o m e n t to m o m e n t it is selecting modules according to its interests, the phenomena of attention, and is itself integrating from all this diversity to give the unified conscious experience. (HM 229)

Four flaws in Eccles's conception

T h e m e t a p h o r s are striking, a n d have echoes in current . ^^.c t U _ 2o neuroscientific theory. 2 0 Nevertheless, Sperry's discoveries

So, for example, Crick called his theory of attention 'the searchlight hypothesis', since, he claimed, the reticular complex and the pulvinar promote only a small proportion of the activity of the thalamus on its way to the cortex, and this activity can be likened to a searchlight that lights up a part of the cortex. Crick suggested that the thalamic reticular complex and the pulvinar interact with the brain stem and with cortical mechanisms to reach a salient decision as to which neuronal groups that are active will be 'brought into consciousness' by the spotlight of attention (F. Crick, Function of the thalamic reticular complex: the searchlight hypothesis', Proceedings of the Natiional Acadeny of Science USA, 81 (1984), pp. 4586-5490). Similarly, the notion of a scanning device or monitor' in the brain has been invoked by Weiskrantz in connection with his investigations of blind-sight. In his view, the awareness that a normally sighted person has of whether he sees something in his visual fieid and of what he sees results from the operation of a neural monitoring system. Conscious experience, according to Weiskrantz, is the product of the monitoring function of the brain (L. Weiskrantz, 'Neuropsychology and the nature of consciousness', in C. Blakemore and S. Greenfield (eds), Mindwaves (Biackwell, Oxford, 1987), pp. 307-20). It is interesting that whereas Crick and Weiskrantz apply these metaphors to the brain, Eccles applied them to the mind.

56

Philosophical Problems in Neuroscience

have none of the dramatic impBcations that Eccles imputed to them. There are four fla^ in Eccles's conception to which we wish to draw attention. First, the phenomena were misdescribed. it is not just thp 1. The phenomena resultant upon , . .. , . . , . , , „ , , . , . , .. , neural activities ofr the right hemisphere that are unknown hemtspherectomy were misdescribed _ ,. „ , ° .. . , , to the subject - all the activities of the brain are unknowi to subjects, who do not, after all, perceive their own brains (and, even if they could, do no have electron microscopes for eyes). It is true that the right hemisphere cannot 'exprés itself in language', any more than the right leg — because there is no such thing as a part o a human being expressing itself in language (see §§3.1-3.4). So the left hemisphere canno 'express itself in language' either. The right hemisphere is not able 'to disclose an1 experience of consciousness' that we can recognize, because there is no such thing as subordinate part of a person being conscious. As will be argued in detail in chapter 3, it i only human beings (and other animals) who are conscious {or unconscious), and consciou of (or not conscious of) various things — not their subordinate parts. The left hemispher is equally lacking in 'any experience of consciousness'. Finally, the left hemisphere has n< 'verbal and ideational abilities', although the verbal and ideational abilities of normal humai beings are causally dependent upon the normal functioning of the left hemisphere. 21 _ „_, . ,. , Secondly, the so-called self-conscious mind is not ai 2. I he self-conscious mind is .. , , , , . , . r ,. , entity ofc anyJ kind, but a capacity otr human beinss wh< not an entity of any kind , , , have mastered a reflexive language. They can therefori ascribe experiences to themselves and reflect on the experiences thus ascribed (see §12.6) But the 'self-conscious mind' is not the sort of thing that can intelligibly be said to be 'ii contact with' the brain (let alone with something denominated 'the liaison brain'). _ , , Thirdly, Eccles's main hypothesis is unintelligible. If th j . Incoherence in Eccles's ,r . . . .,., t . , , , . self-conscious mind were, per impossible, actively engage in reading out' from areas in the dominant hemispher and 'selecting from these modules according to attention', then the self-conscious min would have to perceive or be aware of the neural modules in question (otherwise ho\ could it 'read them out'?), and know which ones to select for its purpose (otherwise th wrong ones might constantly be selected). Or, to put matters more lucidly, for any of flu story to make sense, human beings would have to be aware of the neural structures an operations in question, and, from moment to moment, decide which ones direcdy t< activate, and, of course, have the capacity to do so. But we possess no such knowledg and no such capacity. Finally, it is confused to suppose that the raison d'etre c 4. The very notion of the self- , i , r . . ,, . , , r , . , , . the self-conscious mind is to engender the unity of th J consaous mind presupposes the unity self and, as our contemporaries would put it, solve th nc ev„Bv:„„,„ L Of £*\jjcrlCiíCc

binding problem'. For any talk of a person, or of a being as having a mind already presupposes the unity of experience and cannot be invokei to explain it. Neuroscientists' misdescriptions of split-brain patients' abilities and their exercise is examined ani rectified in §14.3 below.

Sherrington and his Proteges

57

Eccles's dualism was misconceived. Contemporary Eccte's errors can J; ' neuroscientists are eager to dissociate themselves from his ¡ubsli""M doctrines and to dismiss his ideas as silly. This is mis¡ ,•_., 0f the self-conscious mina ce l J P guided. Eccles had the courage to face difficult problems j t o pursue his ideas about them to their logical conclusions. That his ideas are wrong • t r l i e j ¿xid much can be learnt from the errors in question. It is, however, a sad mark of how Httle many neuroscientists have learnt from Eccles's struggles that they apparently believe that the problems that Eccles's interactionist dualism was designed to answer can be solved by substituting the brain for Eccles's 'self-conscious mind'. Problems regarding how the mind can bring about movements of the muscles and limbs by acts of will are not solved by supposing, as Libet does, that it is the brain that decides what muscles and limbs to move. Although it is misguided to suppose that the mind is in liaison with the left hemisphere, it is no less misconceived to suppose, as do Sperry, Gazzaniga and Crick, that the hemispheres of the brain know things, have beliefs, think and guess, hear and see. For these are functions of human beings and other animals, not of brains or half-brains (which enable human beings to exercise those functions). And, as we have noted, although it is confused to suppose that the mind scans the brain, it is equally confused to suppose that the brain must scan itself in order to generate awareness or self-consciousness — as if it lay in the nature of self-consciousness that it necessarily involves a self-scanning process, if not of the mind, then of the brain. In short, the lessons that can be learnt from Eccles's failure have largely yet to be learnt. We shall endeavour to show this in some detail in later chapters.

2.4 Wilder Penfield and the 'Highest Brain M e c h a n i s m ' Wilder Penfield (1891-1976) was born in Spokane, Wash" * ington. After graduating from Princeton in 1913, he won a Rhodes Scholarship to Oxford and entered the School of Physiology there to begin his medical studies under the inspiring influence of Sherrington. He followed Sherrington's interest in histology and, in particular, in neurocytology. After obtaining his BA in physiology at Oxford, he went to the Johns Hopkins Medical School, where he finished his medical degree in 1918. His first research concerned changes in the Golgi apparatus of neurons after axonal section. In 1924 he began to study the healing processes of surgical wounds in the brain. On Sherrington's advice, he spent some time in Madrid working with Pio del Rio-Hortega, learning to use the histological methods of his briUiant teacher Ramón y Cajal. To this end, surgical specimens of brain scars were collected from patients who had been operated on for post-traumatic epilepsy. Penfield was aware of the studies on cortical localization Pefiheid s achievement . , . , . , _, , , . , in the primate bram that Sherrington had earned out, and which have been described above. In 1928, he went to Breslau to work with Otfrid Foerster, to learn his method of gentle electrical stimulation of the cortex of epileptic patients while they were under local anaesthesia during the excision of epileptogenic scar tissue. During these procedures he learnt the method of operating under local anaesthesia,

58

Philosophical Problems in Neuroscience

using electrical stimulation to identify the sensory and motor cortex to guide the surgiCai excision. This technique was to be used to singular effect by Penfield in Montreal, whore in 1934, he established the famous Montreal Neurological Institute at McGill University which was devoted to the study and surgical treatment of focal epilepsy. Such stimulation made it possible to locate exactly the position of the sensorimotor cortex and of the cortex subserving speech, so that these vital areas could be spared during the surgical excision. I n some instances the stimulation might activate the more excitable epileptogenic cortex and reproduce a portion of the patient's habitual seizure pattern. This enabled the surgeon to identify the site of the physiologically deranged epileptic focus. Penfieid's mastery of these procedures was subsequently summarized in a series of monographs on brain surgery for epilepsy. Penfield noted in 1938 that stimulation of certain parts of the temporal cortex l n patients occasionally excited the vivid recall of previous experiences. It became evident that almost half of the patients afflicted with epilepsy had seizures that could be shown to originate in one or other of the temporal lobes. This work on temporal lobe epilepsy led to very important observations regarding the hippocampus and memory function, a? well as the localization of the cortex subserving the latter. By 1951 Penfield, together with Milner, had shown that removal of one hippocampus on the medial aspect of the temporal lobe resulted in severe memory disorder in patients who were later found to have damage to the hippocampus on the opposite side. Thus the bilateral loss of function of the hippocampus led to the complete inability of these patients to remember any postoperative occurrence. This memory loss was not accompanied by any loss of intelligente or attentive capacity. Penfieid's analyses of the electrical stimulation of the coriex of 1,132 conscious patients undergoing brain surgery greatly extended our knowledge of functional localization, especially with regard to memory and to that most human of capacities, speech. „ r ,,, , , , , Penfield s methodological commitment

Already in his student days, Penfield had had a 'sense of , , . , . . , , . ,, , , wonder and a profound curiosity about the mind . \\v hen r

he turned from the study of the animal to the hiinun brain, his 'planned objective', he later wrote, was 'to come to understand the mechanisms of the human brain and to discover whether, and perhaps how, these mechanisms account for what the mind does'. 22 Studying under Sherrington, he came to 'the realization that the brain was an undiscovered country in which the mystery of the mind of man might some day be explained'. He was, of course, fully aware of Sherrington's views on the relation of mind and brain. In the final paragraph of the foreword to his great book 7 k Integrative Action of the Nervous System (1906), Sherrington had remarked, 'That our being should consist of two fundamental elements offers, I suppose, no greater inherent impi «liability than that it should rest on one only'. Penfield, however, took the view that the neuroscientist should endeavour to explain the behaviour of animals, including human:*,

22

W. Penfield, The Mystery of the Mind: A Critical Study of Consciousness and the Human lit.i (Princeton University Press, Princeton, 1975), p. 1. Subsequent references in the text to this voninn* will be flagged 'MM.

Sherrington and his Proteges

,

59

the basis of n e u r o n a l mechanisms alone. O n l y if that failed, h e t h o u g h t , should o n e to alternative forms of explanation. A n d t h r o u g h o u t his career as a n e u r o reeon, he retained this m e t h o d o l o g i c a l c o m m i t m e n t . e recourse

penjieia o

T o w a r d s the e n d of a l o n g life dedicated to neurosurgery neurology, Penfield published a small v o l u m e entitled

ancj

The Mystery of the Mind. This was, h e w r o t e , ' t h e final report of m y experience' — an overview of w h a t h e h a d achieved in respect of his youthful objective. ' T h e nature of the mind', h e averred, 'presents the fundamental p r o b l e m , perhaps the most difficult a n d m o s t important of all p r o b l e m s ' (MM 85). W h a t h e w i s h e d at last to d o , h e w r o t e in the orefece, w a s to 'consider the evidence as it stands, and ask the question Do brain mechanisms account for the mind? C a n the m i n d b e explained b y w h a t is n o w k n o w n a b o u t the brain?' (MM, P- xii')- R e f e r r i n g explicitly to the a b o v e - q u o t e d r e m a r k of Sherrington's, Penfield judged that 'the t i m e has c o m e to l o o k at his t w o hypotheses, his t w o "improbabilities". Either brain action explains t h e m i n d , or w e m u s t deal w i t h t w o elements' (MM 4). Despite his methodological c o m m i t m e n t , Penfield found himself driven towards a Cartesian view not unlike that of his great teacher. ' F o r m y o w n part', h e w r o t e , after years of striving to explain the mind o n the basis of brain-action alone, I have come to the conclusion that it is simpler (and far easier to be logical) if one adopts the hypothesis that our being does consist of two fundamental elements. . . . Because it seems to me certain that it will always be quite impossible to explain the mind o n the basis of neuronal action within the brain, and because it seems to me that the mind develops and matures independendy throughout an individual's life as though it were a continuing element, and because a computer (which the brain is) must b e programmed and operated by an agency capable of independent understanding, I am forced to choose the proposition that our being is to be explained on the basis of two fundamental elements. This, to my nund, offers the greatest likelihood of leading us to the final understanding toward which so many stalwart scientists strive. (MM 80) What led h i m to this conclusion? T w o features in particular had impressed Penfield. First, given his specialization in epilepsy cases, h e was, unsurprisingly, impressed b y the p h e n o m e n a of epileptic automatism. Second, h e was powerfully struck b y the responses elicited from patients in reaction to electrode stimulation d u r i n g surgery. Penfield s interpretation of ., . . epileptic automatism

A patient, suffering an epileptic seizure that has i n d u c e d ... r . automatism, will often continue to execute whatever m o r e

or less stereotypical tasks h e was engaged in. H e will, however, b e in a fugue c o n d i t i o n — that is, after recovery h e will r e m e m b e r n o t h i n g of what he has d o n e during the seizure. Penfield interpreted automatism as s h o w i n g that the epileptic seizure disconnected the m i n d from w h a t , following Hughlings Jackson, 2 3 h e called 'the brain's highest m e c h a n i s m ' (a precursor of Eccies's 'liaison brain'). H e t o o k it that the brain, during the p e r i o d o f automatism, is controlling the b e h a v i o u r of a ' h u m a n

J. H. Jackson, ' O n the anatomical, physiological and pathological investigations of epilepsies', West Riding Lunatic Asylum Medical Reports, 3 (1873), pp. 315-19.

60

Philosophical Problems in Neuroscience

a u t o m a t o n ' in accordance w i t h antecedent ' p r o g r a m m i n g ' b y the m i n d . Just as the prog r a m m i n g of a c o m p u t e r comes 'from w i t h o u t ' , so t o o the p r o g r a m m i n g of the brain w h i c h is, Penfield claimed, a biological c o m p u t e r , is effected b y the m i n d via the brain's highest mechanism. P u r p o s e comes to it from outside its o w n mechanisms. Short-terra p r o g r a m m i n g obviously serves a useful purpose, m a k i n g possible automatic continuation of routine tasks, and this is visibly a n d strikingly manifest during periods of such epileptic seizures. That this highest mechanism, most closely related to the mind, is a truly functional unit is proven by the fact that epileptic discharge in gray matter that forms a part of its circuits, interferes with its action selectively. During epileptic interference with the function of this gray m a t t e r . . . consciousness vanishes, and with it goes the direction and planning of behaviour. That is to say, the mind goes out of action and comes into action with the normal functioning of this mechanism. T h e human automaton, which replaces m a n w h e n the highest brain-mechanism is inactivated, is a thing without the capacity to make completely new decisions. It is a thing without the capacity to form n e w m e m o r y records and a thing without that indefinable attribute, a sense of humour. T h e automaton is incapable of thrilling to the beauty of a sunset or of experiencing contentment, happiness, love, compassion. These, like all awarenesses, are functions of the mind. T h e automaton is a thing that makes use of reflexes and skills, inborn and acquired, that are housed in the computer. (MM 47)

T h o u g h Penfield v e n t u r e d n o testable hypotheses a b o u t h o w this interaction occurs, h e claimed that the highest brain m e c h a n i s m is, as it w e r e , the m i n d ' s executive. It accepts directions from the m i n d , and passes t h e m o n to the various mechanisms of the brain (MM 84). T h e m i n d directs the brain in action. It has n o m e m o r y of its o w n . But the contents of the stream of consciousness are r e c o r d e d in the brain (as seems evident from the inadvertent retrieval of long-lost m e m o r i e s d u r i n g cortical stimulation of the brain during operations). So, w h e n the m i n d needs to retrieve a m e m o r y , in a flash it opens the files of r e m e m b r a n c e in the brain t h r o u g h the highest brain mechanism (MM 49). Reflection o n s o m e o f the p h e n o m e n a c o n s e q u e n t o n . , . , . , . . , in r u corneal stimulation during operations led Pentield to sim, - , , - , , , £ conclusions. So, for example, a patient, w h o s e speech cortex' was interfered w i t h b y an electrode, exhibited exasperation w h e n h e could n o t identify a picture of a butterfly. O n w i t h d r a w a l of the electrode, h e said, ' " N o w I can talk. Butterfly. I c o u l d n ' t get t h e w o r d "Butterfly", so I tried to get the w o r d " m o t h " ! ' It is interesting to see h o w Penfield construed and explained this temporary i m p a i r m e n t of the patient's n o r m a l identificatory capacities.

Penfield s interpretation ofphenom. , . , ena consequent on electrode cortical stimulation

It is clear that while the meaning of the corresponding word. for a second time to

the speech mechanism was temporarily blocked he could perceive picture of a butterfly. H e made a conscious effort to 'get' the Then, not understanding w h y he could not do so, he turned back the interpretative m e c h a n i s m . . . and found a second concept that

Sherrington and his Proteges

61

he considered the closest thing to a butterfly. 24 H e must then have presented that to the speech mechanism, only to draw another blank. (MM 52) According to Penfield, concepts are stored away in the mind's c o n c e p t m e c h a n i s m in the brain, from w h i c h the m i n d selects the c o n c e p t it requires. T h a t c o n c e p t is t h e n presented in the stream of consciousness, a n d if the m i n d approves of the selection, the highest brain mechanism flashes this n o n - v e r b a l c o n c e p t to t h e speech mechanism, w h i c h , w h e n functioning normally, will present to the m i n d the w o r d that is appropriate for the c o n c e p t (MM 53). penfield was equally impressed b y t h e fact that w h e n neural stimulation t o t h e brain caused a hand m o v e m e n t , the patient invariably responded, 'I didn't d o that. Y o u did.' And equally, w h e n cortical stimulation caused vocalization, the patient said, 'I d i d n ' t m a k e that noise. Y o u pulled it o u t of m e . ' It was striking that n o form of electrical stimulation to the cortex could i n d u c e a patient to believe or to decide (MM 77). It is n o t surprising that Penfield d r e w the conclusion that belief a n d volition are functions o f the m i n d . . , , . . . .. The mind and its interaction with "., , . . ,,, ,. , . , . the brain via the highest brain umism>

A man's mind, Penfield concluded, is the person ( M M 6 1 ) . . . . . . „ , . ' it is the m i n d that is aware of w h a t is g o i n g o n , that ° ° reasons and decides, and understands (MM 7 5 f ) . T h e person walks a b o u t the w o r l d [sic], d e p e n d i n g always upon his 'private c o m p u t e r ' (i.e. his brain), w h i c h h e programs continuously (MM 61). The highest brain m e c h a n i s m is the m e e t i n g place of m i n d and brain, the psychophysical frontier (MM 53). T h e m i n d , in m a k i n g decisions, causes the highest brain m e c h a n i s m to send neuronal messages to o t h e r mechanisms in the brain, and data stored in the brain are admitted to consciousness. H o w is interaction effected? H e r e Penfield speculated that there must be a second form of energy (other than the electrical energy that is used by the highest brain function to innervate t h e n e r v o u s system) w h i c h is available to the m i n d . This, he conjectured, must b e m a d e available to the m i n d in its w a k i n g hours by t h e highest brain m e c h a n i s m . The mind vanishes w h e n the highest brain mechanism ceases to function due to injury or due to epileptic interference or anaesthetic drug. M o r e than that, the mind vanishes during deep sleep. What happens w h e n the mind vanishes? There are two obvious answers to that question; they arise from Sherrington's two alternatives - whether man's being is to be explained on the basis of one or two elements. (MM 81)

Penfield t h o u g h t it preposterous to suppose that the m i n d is merely a function of the brain, and so ceases to exist w h e n it 'vanishes' in sleep or epileptic automatism a n d is re-created afresh each t i m e the highest brain m e c h a n i s m functions normally. R a t h e r , h e concluded, the m i n d is 'a basic element', and has a ' c o n t i n u i n g existence'. ' O n e must assume', he w r o t e , 'that a l t h o u g h the m i n d is silent, w h e n it n o longer has its special

renheld obviously meant that it was the closest approximation to the concept of a butterfly.

62

Philosophical Problems in Neurosdence

c o n n e c t i o n to the brain, it exists in the silent intervals and takes o v e r control w h e n the highest brain m e c h a n i s m goes i n t o action' (MM 81). So, t h e highest brain mechanism switches off the p o w e r that energizes the m i n d w h e n e v e r o n e goes to sleep, a n d switches it o n again w h e n o n e awakens. Is the explanation improbable?, Penfield queried. It is not so improbable, to my mind, as is the alternative expectation [explanation] - that the highest brain mechanism should itself understand, and reason, and direct voluntary action, and decide w h e r e attention should be turned and what the computer must learn and record, and reveal o n demand. (MM 82) Penfield's neo-Cartesianism is n o advance over that of „, , _ , „ , .c „ , ,. J Sherrington and Eccles. B u t if w e are to learn anything from his errors, w e must n o t simply dismiss t h e m as misguided and m o v e o n to other matters. T h a t will merely ensure that w e learn n o t h i n g from his endeavours. W e must ask w h a t w e n t w r o n g , w h a t drove o n e of the greatest neurosurgeons a n d neurologists of all times to embrace such a misconceived v i e w of the m i n d a n d brain? It should b e n o t e d that there are at least t w o fundamental . —, v , P\ . , , presuppositions that Penfield shared w i t h Sherrington and 1. ihe Cartesian conception of the r , r rm, . _, _ , . , . , Eccles. T h e first was a Cartesian c o n c e p t i o n ot the mind. c mina Like Descartes, Penfield c o n c e i v e d of the m i n d as an i n d e p e n d e n t substance (or, as h e puts it, 'a fundamental e l e m e n t ' that has 'continuing existence'). Like Descartes, h e identified the p e r s o n w i t h the m i n d , rather than w i t h the living h u m a n being. Like Descartes, h e t o o k t h e m i n d t o b e t h e bearer o f psychological attributes, 2 5 and consequently conceived of h u m a n beings as subjects of psychological predicates only derivatively. A n d like Descartes, h e t o o k the m i n d to b e a causal agent that can b r i n g a b o u t changes i n the b o d y by its actions. T h e second presupposition is that the question w h i c h so 2. The assumption that the question i , j, i i • , i u u u , , / . , . deeply disturbed h i m - namely, w h e t h e r brain m e c h a n r ic/ whether brain mechanisms can . , , . , ,£ A • , • • • i isms a c c o u n t for the m i n d , w h e t h e r the m i n d can be account jor the mind is an empirical ' explained b y reference to w h a t is k n o w n a b o u t the brain om - is an empirical question. Like Sherrington, Penfield c o n c e i v e d of t h e matter as a c h o i c e b e t w e e n t w o different empirical hypotheses. E i t h e r w e can explain everything the m i n d does — for example, thinks a n d believes, reasons and concludes, has wants, forms intentions a n d purposes, a n d decides to act — b y reference to neural states and events, or w e must conceive of the m i n d as an i n d e p e n d e n t substance in i m m e d i a t e causal interaction w i t h the brain, a n d h e n c e w i t h the b o d y . T h e choice b e t w e e n these t w o hypotheses is to b e d e t e r m i n e d b y the evidence that supports t h e m severally a n d by their relative explanatory p o w e r s . B o t h presuppositions are misconceived. T h e m i n d , as w e have already intimated, is not a substance o f any kind. T a l k o f t h e m i n d is m e r e l y 2. fagon de parley for talk a b o u t h u m a n 25 Indeed, to explain what the mind or spirit is, Penfield quoted Webster's Dictionary: 'the e l e m e n t . . . in an individual that feels, perceives, thinks, wills and especially reasons' (MM 11). Penfield s neo-Cartesianism

Sherrington and his Proteges

63

. powers and their exercise. We say of a creature (pnmarcn r ,1, Cntiásms qj Penfield s presupposi- ., , . , . , ........ . ,,. ,. , " ,, ily oí J a human being) that it has a mind if it has a certain tiotts: 1. misconceptions about the . c ,„ • i range of active and passive powers of intellect and will in particular, conceptual powers of a language-user that make sell-awareness and self-reflection possible. The idioms that involve the noun 'mind' have as their focal points thought, memory and will. And they are all readily paraphrasable into psychological expressions in which the word does not occur (we shall discuss this matter in some detail in §3.10). A person is not identical with his mind. A mind is something (but not some thing) a person is said to have, not to be. In having a mind, an animal (that is thereby also a person, and a bearer of rights and duties) has a distinctive range of capacities. And it is obvious both that an animal cannot be identical with an array of capacities, and that if a human being loses enough of those distinctive capacities, he can cease to be a person (and exist only in a Vegetal state'). It is not the mind that is the subject of psychological attributes, any more than it is the brain. It is the hving human being - the whole animal, not one of its parts or a subset of its powers. It is not my mind that makes up its mind or decides; it is not my mind that calls something to mind and recollects; and it is not my mind that turns its niind to something or other and thinks — it is I, this person. Hence, too, the mind is not a causal agent that brings about changes in the body and its limbs by its actions. On the contrary, it is human beings that deliberate, decide and act, not their minds. „ „ , , , . , Consequently, Penfield's second presupposition is mis2. Whether the brain can account . , , wr. . , , - , • , , . , , , , j . ^ . , , guided. Whether we can account for the mind m terms for the mind is not an empirical ° auation brain alone, or must account for the (supposed) activities of the mind (e.g. thought, reasoning, wants and purposes, intentions and decisions, voluntary and intentional actions) by reference to the mind itself, conceived of as an independent substance and therefore causal agent, is not a matter of choice between two empirical hypotheses. If these were empirical hypotheses, then either could in principle be true; that is, both would present intelligible possibilities, and it would be a matter of empirical investigation to discover which is actually the case. But that is not how it is at all. . . . , . , First, it is not the mind that thinks and reasons, wants It is neither the brain nor the mind ,. . , . . . , , ,i , . ¿1 i• , r , , • i things and has purposes, forms intentions and makes & that is the subject of psychological r r > attributes decisions, acts voluntarily or intentionally. It is the human being. We do indeed characterize a person as having a clear, rigorous or decisive mind. But these are merely ways of characterizing the person's dispositions in respect of thought and will. If we want to understand why a normal person reasoned the way he did, thought what he did, has the goals and purposes that he has, and why he decided as he did, formed such-and-such intentions and plans, and acted intentionally, no neurological account will clarify for us what we wish to be clarified. To this extent Penfield was right. Where he was wrong was in the supposition that what we need is an explanation in terms of the activities of the person's mind — where the latter is conceived of as an agent with causal powers. Rather, what we want is an explanation in terms of the person's reasoning, hence too by reference to what he knew or believed, and, in the case of practical reasoning, by reference to his goals and purposes. And if our

64

Philosophical Problems in Neuroscience

explanation renders his reasoning intelligible, no further information about neural events in his brain can add anything. All a neural explanation could do would be to explain how ^ it was possible for the person to reason cogently at all (i.e. what neural formations must be in place to endow a human being with such-and-such intellectual and volitional capacities) 'but it cannot rehearse the reasoning, let alone explain its cogency. Similarly, if we are puzzled by a person's actions, if we Neither the causal agency of the w l .s n, t o , , . . , , /-,,,,,, c , , r , know why A signed a cheque for 7,200, nn brain nor the causal agency of the . . ° . ,., , . , , - - , , • ' > tanswer in terms or brain functions is likely to satisfy ¡R 7 mind explains intentional action . . "• h We want to know whether A was discharging a debt * making a purchase, donating money to charity, or betting on a horse — and once we know " which of these is the case, we may also want to know what A's reasons were. A description of neural events in A's brain could not possibly explain to us what we want to ' have explained. If we wish to know why A caught the 8.15 a.m. to Paris, a description of neural events cannot in principle satisfy our need for an explanation. But the answer that í he had a committee meeting there at 2.00 p.m. that was to decide upon such-and-such a < project for which A is responsible may give quietus to our curiosity. If A has murdered B, . we may wish to know why. We may be given a reason, and still remain dissatisfied, wishing to understand more — but the 'more' we wish to understand is most probably A's motive, not what neural events occurred at the time of the killing. We want to know „' whether he killed B out of revenge or out of jealousy, for example, and that requires a u quite different narrative from anything that neuroscientific investigation could produce. Explanation of action by redescription, by citing agential reasons, or by specifying the agent's motives (and there are other forms of explanation of related kinds) are not replaceable, even in principle, by explanations in terms of neural events in the brain. This is not an , empirical matter at all, but a logical or conceptual one. The type of explanation is categorially different, and explanations in terms of agential reasons and motives, goals, and purposes, are not reducible to explanations of muscular contractions produced as a consequence of neural events (see chapter 13). But equally, such explanations are not couched in terms of the activities of the mind, conceived as an independent substance with causal powers of its own. In this sense, Penfield's dilemma is a bogus one. He was perfectly right to think that one cannot account/for human behaviour and experience in terms of the brain alone, but wrong to suppose that the idea that one might be able to do so is an intelligible empirical hypothesis as opposed to a conceptual confusion. He was also wrong to suppose that the alternative is accounting for human behaviour and experience in terms of the causal agency , of the mind, and wrong again in thinking that that too is an empirical hypothesis There '< is no need whatsoever to impale oneself on either of the horns of Penfield's dilemma. Once these presuppositions are jettisoned, it becomes The hypothesis that mind—brain . , , , . , , , . , . , . , easier to see why the explanation ofr human behaviour in J interaction can explain human beha. • , , - , r , . , . „ . , , terms of the interaction or the mind (conceived as an vwur is logically incoherent . . . independent substance) and the brain is misconceived. It is not a false empirical hypothesis, but a conceptual confusion. For inasmuch as the mind is not a substance, indeed not an entity of any kind, it is not logically possible for the mind to function as a causal agent that brings about changes by acting on the brain. This is not an empirical discovery, but a conceptual clarification. (But it is equally mistaken to

Sherrington and his Proteges

65

ose that substituting the brain for the Cartesian mind is any less confused. That too is „ empirical hypothesis, but a conceptual muddle, which likewise stands in need of concepta 1 clarification.) Consequently, Penfield was mistaken to think that what Neither epitep u, . so impressed him - namely, the phenomena of epileptic r , i„Wf stimulation of the brain . , , • V , , • eteaw automatism and the various tacts that characterize electrode SH PP° stimulation of the brain - constitute empirical support r a dualist hypothesis. Epileptic automatism does not show that the mind has become disconnected from the 'highest brain mechanism' to which it is normally connected, and hv which it is supplied with energy of an as yet unknown form.26 What it shows is that during an epileptic seizure, as a consequence of the abnormal excitation of parts of the cortex, the person is temporarily deprived of some of his normal capacities (including memory, the ability to make decisions, emotional sensitivity and a sense of humour), while other capacities, in particular capacities for routine actions, are retained. The phenomena are indeed striking, but they amount to a dissociation of capacities that are normally associated, not to a disconnection of substances that are normally connected. They do not show that the brain is a computer or that the mind is its programmer. The brain is no more a computer than it is a central telephone exchange (the previously favoured analogy), and the mind is no more a computer programmer than it is a telephonist. It is perfectly true that the capacity to continue routine tasks unreflectively is useful (and the expression 'short-term programming' is an apt metaphor here). It is also true that the purposes pursued by a person are not the purposes of the person's brain. But it does not follow from this that they are the purposes of the person's mind. They are the purposes of the person — and they are to be understood in terms of facts about human life, social forms of life, antecedent events, current circumstances, agential beliefs and values, and so forth, not in terms of neural events and mechanisms. But it is, of course, true that, but for the normal functioning of the brain, a human being would not have, and would not be able to pursue, the normal kinds of purposes that we do pursue. The various phenomena that characterize electrode stimulation of the brain are similarly misconstrued by Penfield. The case of interference with the 'speech cortex' does not show that there is any such thing as a 'concept mechanism' in the brain that stores non-verbal concepts that can be selected by the mind and then presented to the speech mechanism to be matched to the word that represents the concept. That is picturesque mythology, not an empirical theory. Words are not names of concepts, and do not stand for concepts, but rather express them. Concepts are abstractions from the use of words. The concept of a cat is what is common to the use of 'cat', 'chat', 'Katze', etc. The common features

It is striking to compare Penfield's conception of this matter with Descartes's remarkable simile in his Treatise on Man: 'when a rational sou! is present in this machine [namely, the body] it will have its principal seat in the brain, and reside there like the fountain keeper who must be stationed at the tanks to which the fountain's pipes return if he wants to produce, or prevent, or change their movements in some way' (AT XI, 131). Here the tank is the ventricle in which the pineal gland is allegedly suspended, the pipes are the nerves, and the water the animal spirits.

66

Philosophical Problems in Neuroscience

of the use of these words is not something that can be stored in the brain or anywhere else independently of a word (or symbol) that expresses the concept. The patient whom Penfield describes could not think of the word 'butterfly' with which to identify the object in the picture presented to him. He knew that the object belonged to a class which resembles a different class of insects (viz. moths), and tried, equally unsuccessfully, to tliint of the word for members of the second class. This temporary incapacity is iucotrculv described as knowing the concept but being unable to remember the word for it. The supposition that the mind might be presented with non-verbal concepts from which to choose presupposes that there is some way of identifying non-verbal concepts and distinguishing one from the other independently of any words or symbols that express ihcm, But that makes no sense. It is certainly interesting that Penfield found that electrode stimulation could not induceeither belief or decision. But this does not show that believing and deciding are actions of the mind, any more than it shows that they are not actions of the brain, ft is true thit thev are not actions of the brain - but that is not an empirical fact that might be shown u> he the case by an experiment. Rather, there is no such thing as the brain's believing or deciding (any more than there is such a thing as checkmate in draughts). But it is also true that they are not actions of the mind either. My mind does not believe or disbelieve anything - / do (although, to be sure, that is no action). Nor does it decide - it is human beings that decide and act on their decisions, not minds. Penfield objected vehemently to the suggestion thai the That the exercise of mental powers ••,. ,. , . . ., r , , . , , , . , mind is a runcaon or the brain, and supposed th.it if it r r is a function of the brain does not , ,;,(!• j , • ,„ were, then the mind would cease to exist during s eep or & show that behaviour and experience < ' are explicable neurally epileptic automatism. The suggestion is unclear, bin one may surely say that the distinctive capacities of intellect and will of a creature that has a mind are a function of the creature's brain (and or other factors too). It does not follow (as Penfield evidently feared it would) that the behaviour and experience of such a creature in the circumstances of life is explicable in neural terim, But nor does it follow that the mind ceases to exist during sleep or epileptic seizure - any more than one's knowledge and beliefs, intentions &nd projects, cease to exist whtn one is asleep. Penfield was rightly impressed by the fact that 'the mind develops and matures independently throughout an individual's life as though it were a continuing element' (MM 80). But he was misled by his unquestioned assumption that the mind is a kind of agent. Had he thought of the mind in more Aristotelian terms as a set of powers or capacities, he would have been closer to the truth, and less prone to conceptual illusion. For the continuous possession of capacities is not interrupted by sleep or even by epileptic automatism, even though the agent cannot exercise some of his normal capacities during the seizure. And the developing unity of a person's mind is not the development of a substance distinct from the human being himself, but rather the emergence of a determinate character and personality, an intellect with certain distinctive characteristics and a will with a coherent array of preferences - all of which are traits of the person. Penfield thought that a form of Cartesian dualism is more probably correct than w-hat he conceived to be the alternative: namely, ascribing understanding, reasoning, volition and voluntary action, as well as deciding, to the brain itself. It is very striking and

Sherrington and his Proteges

1ttit Penfield thought the less •u ' assuages his injured limb, limps, or nurses his injury, grimaces and exclaims 'Ow', 'It hurts', or 'I have a pain', we take such pain-behaviour in these circumstances to be justifying grounds or evidence for ascribing pain to the person. This does not mean that we normally infer that a person is in pain from such observed evidence. We typically recognize immediately that a person has hurt himself, we see and hear that he is in pain, can see the pain in his face. But it is the person's pain-behaviour that warrants our immediate pain-ascription, and it is by reference to such evidence that we might answer the question 'What grounds did you have for taking him to he in pain?' Similarly, we say of an animal or a human being that they perceive something in their field of perception if, for example, they respond to what is visible (or audible, etc.) in appropriate ways. So, evidence that a dog sees a cat is that it responds to what is visible to it - for example, follows the cat with its eyes, displays interest in the visible behaviour of the cat, chases it and responds to its movements by adjusting its pursuit to the visible twists and turns of the fleeing cat. Likewise, we ascribe a given belief to a human being if, for example, he asserts that such-and-such is the case, or avers that he believes that things are There are other methodological objections that have been elaborated by Quinean philosophers of science. They carry weightier philosophical baggage, and will be considered separately, in §14.1. Readers who would like to examine our further arguments may wish to jump forward.

82

t h u s - a n d - s o , or if reasons that things b e so. W e d o n o t normally take h i m h i m this belief are

Philosophical Problems in Neuroscience

h e acts and explains his d o i n g w h a t h e is d o i n g b y reference to the are thus-and-so — that is, by reference to w h a t h e k n o w s or believes to infer that A believes s u c h - a n d - s u c h from the fact that h e said so, we to believe w h a t h e says h e believes; b u t o u r grounds for ascribing ic that h e said w h a t h e said.

_, , , The evidential grounds for the , , , . , . , ascription of psychological attributes to others are not inductive, but rather criterial; i.e. the evidence is logically good evidence

T h e primary grounds or evidence for the ascription of , 1 - 1 1i_ 1 L 1 1 • psychological predicates to a n o t h e r are behavioural, h is r } • r £he b e h a v i o u r °f that b™£ i n appropriate c i r c u m s t a n c e . This ' h o w e v e r , is n o t inductive evidence. Inductive evidence *s d e t e r m i n e d by correlation of c o n c o m i t a n t p h e n o m e n a . So it presupposes n o n - i n d u c t i v e identification of the p h e n o m e n a that are observed to be correlated as a matter of fact. B u t pain a n d pain-behaviour are n o t correlated as a matter of brute fact. It is n o t an empirical discovery that w h e n people are in pain, they groan, cry o u t and assuage their injury. N o r is it an intelligible possibility that pain m i g h t systematically b e correlated w i t h smiling a n d laughing, j - , opposed to b e i n g correlated w i t h crying a n d g r o a n i n g — that is, w i t h pain-behaviour. Similarly, it is n o t an empirical discovery that w h e n a creature sees, it responds to visible objects, uses its eyes to follow t h e m , c a n n o t see w h e n its eyes are closed, or w h e n it is pitch dark. R a t h e r , the primary warrant for the ascription of psychological predicates to anoi her p e r s o n or to an animal is conceptually b o u n d up w i t h the m e a n i n g of t h e relevant predicate. P a i n - b e h a v i o u r is a criterion - that is, logically g o o d evidence for b e i n g i n pain - and perceptual b e h a v i o u r (appropriate to the object p e r c e i v e d a n d to the perceptual modalirv involved) is a criterion for the animal's perceiving. T h a t s u c h - a n d - s u c h kinds of behaviour are criteria for the ascription of such-and-such a psychological predicate is partly constitutive of the m e a n i n g of the predicate in question. 3 5 , , „ Cntenal evidence for the ascription 1 f • 1 -, , • r of psychological attributes to others is defeasible by countervailing evidence

T h o u g h n o n - i n d u c t i v e , the behavioural criteria for js. . , , , . , -, , signing such a psychological attribute to a n o t h e r peí son ,, , -,, are 1,ke 1Ilductlve ' evidence, capable of b e i n g o v e r n d d e n by countervailing evidence. So the support of such evidence does n o t entail the presence of the psychological attribute for w h i c h it is evidence. F o r the logical relation of entailment, as exemplified b y the p r o p o s i t i o n that A is a bachelor entailing the proposition that A is u n m a r r i e d , is n o t defeasible; that is, addition of further propositions cannot u n d e r m i n e the entailment relation. B u t a person may display (apparent) pain-behaviour w i t h o u t b e i n g in pain: h e m a y b e acting o n the si.ige or p r e t e n d i n g to b e in pain in order to deceive. If w e have evidence that h e is doing o n e or the other of these, this evidence defeats the criteria! support given b y his apparent

35

This does not mean that the psychological predication is equivalent in meaning to the behavioiir.il description the truth of which warrants its ascription. It is not only possible but common for people to be in (mild) pain, thinking or intending something, and not exhibit the fact that they are. And it is also possible for people to pretend or dissimulate, i.e. to exhibit such-and-such behaviour, yet not be in such-and-such a psychological state. W e are not defending a form of behaviourism; and tlic conceptual nexus we are insisting upon is an a priori evidential, but not a reductive, one.

The Mereohgical

Fallacy

83

pain-behaviour. A person may say that he thinks or believes something, but he may be lying; he may sincerely aver that he intends to do something or other, but he may be deceiving himself. However, if the criteria for a person's being in pain, believing or intending something are satisfied on an occasion and are not defeated by countervailing evidence in the circumstances, then we are warranted in asserting that he is in pain, does believe or intend. , r . . The criterial grounds for ascribing psychological predicates Tl¡e criterial grounds jor the ascnp, ,. , ., , t . , . , , . to another eperson are conceptually connected with the r tion of a psychological predicate are . . . * , . partly constitutive of the meaning Psychological attribute in question. They are partly conf tl at medicate stitutive of the meaning of the predicate. So the normal ascription of psychological predicates to others does not involve an inductive identification. However, given the possibility of such normal noninductive identifications, the possibility of inductive (non-logical) identification becomes available through inductively correlating subjects' having certain psychological attributes with other phenomena - for example, neurophysiological events in their brain. But any inductive correlation presupposes the criterial nexus that is partly constitutive of the psychological concept in question. Moreover, if such inductive evidence conflicts with the normal criteria for the ascription of a psychological predicate, the criterial evidence overrides the inductive correlation. So, for example, if a person avows that he is not in pain, yet evidence from PET or fMRI suggests that he is, the latter is defeated by the agent's sincere utterance, and the inductive correlations of the data from PET and fMRI with the subject's being in pain need to be re-examined. These evident logical features should give us pause. We The brain does not satisfy the en, -, , , , ,, . r , . .,, ,. do not, outside the neuroscience or psychology laboratory, r tena for being a possible subject oj -> , ascnbe psychological predicates P*111 t o ± e b r a i n " T h o u f h n e u r a l P h e n ° m e n a are well correlated with an animal's or a person's being in pain, the brain does not exhibit pain-behaviour — it does not moan or groan, assuage its broken arm, shed tears or grimace, for there is no such thing as a brain's engaging in such activities. The observed neural phenomena that are concomitants of a person's suffering pain, for example, are not forms of pain-behaviour. They are inductively correlated with being in pain. The correlation is an empirical discovery, which presupposes the concept of pain and its nexus with criterial, non-inductive evidence for the application of the concept of pain to a living creature (not to its brain). Similarly, outside the laboratory, we do not ascribe thinking or believing to the brain, but to the person who thinks or believes, and says what he thinks or believes, or acts, and can be seen to act, on the basis of thinking or believing this or that. We do not observe a brain in a brown study, but Le Penseur sunk in thought. We do not see the brain's credulity, but the person's belief (or disbelief) may be written all over his face as he listens to another's tale. What we can do is correlate a person's thinking of this or that with localized brain activity detected by PET or fMRI. But this does not show that the brain is thinking, reflecting or ruminating; it shows that such-and-such parts of a person's cortex are active when the person is thinking, reflecting or ruminating. (What one sees on the scan is not the brain thinking - there is no such thing as a brain thinking - nor the person thinking — one can see that whenever one looks at someone sunk in thought, but not by

84

Philosophical Problems in Neuroscience

looking at a PET scan - but the computer-generated image of the excitation of cells in his brain that occurs when he is thinking.) Again, this correlation is inductive, not criterial, and it requires antecedent identification of the person's thinking and reflecting by reference to behavioural criteria. It presupposes the concept of thinking, as determined by the behavioural criteria that warrant ascription of thought to a living being. , , . _ , This seems obvious enough on reflection. The moot quesA second explanation of why saent. . . , ,. , , . ., ,„,, , ,¿ .. , , tion is: why are we temptedr to think otherwise? Why have J ists are tempted to ascribe psycholo' so man scientists and gkal attributes to the brain is the V < philosophers) been convinced to array of Cartesian and empiricist t h e contrary? What is the source of the temptation to misconceptions to which many of ascribe consciousness, seeing and hearing, thinking and them are committed believing, feeling and wanting, etc. to the brain? We have conjectured one explanation: namely, that on the rebound from Cartesianism and classical empiricism, scientists and philosophers unthinkingly transposed psychological attributes from the mind to the brain. But there is a deeper explanation, which will shed light on how such questionable transposition could have been executed without question, argument or evidence. Although the ascription of psychological attributes to others rests on (is justified by reference to) behavioural, criterial evidence, the application of psychological predicates to oneself does not. One does not say that one is in pain on the grounds that one is groaning and assuaging one's injured limb. One does not express a belief, admit or confess that one believes this or that on the basis of the evidence of one's own behaviour, and one does not wait to hear what one says in order to find out what one thinks. So how is it that we can say what we are feeling or perceiving, what we think or believe, what we want and intend? It is profoundly tempting to draw the contrast between Misconceptions concerning 'the inner' . , , , , , , . . . . . , ,, , . ,. the mental and the behavioural in a misguided way. and the outer; private ownership . , . , of experience; privileged access; direct S a v i o u r belongs to the public domain, and is eviden as opposed to indirect knowledge of t o a n y o n e appropriately situated to observe it - it is, we m i h t sa the inner; and introspection g Y> t h e 'outer'. By contrast, the mental appears to belong to a special, private domain, accessible only to its subject. It is, we are inclined to think, the 'inner' — and we speak accordingly of people's 'inner (mental) life'. According to this picture (and it is, after all, merely a picture, a metaphor), each person has access to his own mind. W e are further inclined to think of experiences as items owned or possessed by their subject — a temptation fostered by the fact that we speak of having pains, perceptual experiences, thoughts, beliefs, etc. And we are inclined to say such things as 'You can't have my pain' or 'Only I can have my pain'. Thinking thus, we suppose experiences to be essentially privately and inalienably owned — that different people cannot have the same experience, they can have only a similar experience. If so, then each person not only has access to his own mind, he has privileged access to it and to the mental events and processes that go on in it. And this seems obvious, for, to be sure, other people must rely on.my behaviour, on what I do and say, in order to discern what I am feeling or thinking."So, it seems that they know how things are with me indirectly. What they directly perceive is merely outward behaviour. But I have direct access to what is inner, to my own mind. I am conscious of how things are with me.

The Mereologícal Fallacy

85

The faculty whereby I have such direct access to mental states, events and processes is introspection - and it is because I can introspect that I can say how things are with me without observing what I do and say. This array of ideas is fundamental to classical empiricism. Woven into strands of the Cartesian legacy, it is part of the received conception shared by most contemporary neuroscientists. Since it was bound up with Cartesians and British empiricists ascribing psychological attributes to the mind, small wonder that contemporary neuroscientists, sharing this picture but rejecting the conception of the mind as an immaterial substance, should be tempted to ascribe psychological attributes to the brain. According to their conception, the 'access' which introspection gives each person to his own menta! states and processes is a partial and limited access to processes going on in his brain.36

3.4 On the Grounds for Misascribing Psychological Predicates to an Inner Entity This powerful picture, partly Cartesian and partly British Four misconceptions concerning the . . . . , , , , , , . . r , , , . , empiricist, has dominated thought about the mind tor r logical character oj experience and ° . centuries. It fosters a very important misconception of the logical character of .experience and its ascription. We shall clarify" this by spelling out four questionable conceptual commitments of this tempting picture. These are widely accepted among neuroscientists, cognitive scientists and psychologists, as well as some philosophers. , 1. Misconceptions about privacy, , , . ,. concewed as private ownership ofr experience

(1) The mental is a private domain, to be contrasted ., . . ,. , . .. . , , . , . with the rpublic domain of behaviour and physical phe,, , nomena. So, for example, Antomo Damasio holds that r t

consciousness is an 'entirely private, first-person phenomenon which occurs as part of the private, first-person process we call mind'. 37 Edelman and Tononi concur that 'privateness' is 'one of those fundamental aspects of conscious experience that are common to all its manifestations'.38 Stephen Kosslyn and Kevin Ochsner hold that having mental images 'is a quintessential private event', and hence 'notoriously difficult to study' — images being 'internal representations' in the mind of a person.39 The idea is rendered clearer with the assistance of a philosopher, John Searle: 'Subjective e.g., Crick observes that 'There is general agreement that people are not conscious of all the processes going on in their heads . . . While you may be aware of many of the results of perceptual and memory processes, you have only limited access to the processes that produce this awareness' (Astonishing Hypothesis, pp. 19f). A. Damasio, The Feeling of What Happens (Heinemann, London, 1999), p. 12. G. M. Edelman and G. Tononi, Consciousness: How Matter Becomes Imagination (Allen Lane, The Penguin Press, London, 2000), p. 23. S. M. Kosslyn and K. N . Ochsner, 'In search of occipital activation during mental imagery', Trends in Neuroscience, 17 (1994), p. 290.

86

Philosophical Problems in Neuroscience

conscious states have a first-person o n t o l o g y because they exist only w h e n they are experienced b y some h u m a n or animal agent'; furthermore, e v e n if t w o people experience the same p h e n o m e n o n , say a concert, each person's experience is distinct from each o t h e r person's experience. T h e y m a y b e qualitatively identical (as y o u r p e n n y may b e qualitatively identical w i t h m i n e ) , b u t t h e y are nevertheless numerically distinct. 4 " As the great logician G o t t l o b Frege wittily put it, y o u can't have m y pain, and I can't have y o u r sympathy. 4 1 Every experience, every conscious state, needs an o w n e r , and each experience or mental state has only o n e o w n e r . T h e m e n t a l is privately and inalienably owned. (2) T h i s private d o m a i n can b e a p p r e h e n d e d by means , . , . , . ,_ , - ^ c of introspection, w h i c h is a faculty akin to perception. .„. t ,, , . t r William JTames held that introspection means, oí course, r . ' the l o o k i n g i n t o o n e s o w n m i n d a n d reporting there w h a t w e discover. Everyone agrees that we there discover states of consciousness.'* Introspection is sometimes held to be the only reliable indicator o f the o c c u r r e n c e a n d character of the inner, conceived as conscious experience. A c c o r d i n g to Benjamin Libet, 'any behavioural evidence w h i c h does n o t require a c o n v i n c i n g introspective r e p o r t c a n n o t b e assumed to b e an indicator of conscious, subjective experience'. 4 3 Nicholas H u m p h r e y , in a similar vein, has claimed that 'conscious experience is t h e set of subjective feelings w h i c h , at any o n e time, are available to introspection'; 4 4 elsewhere h e has c o m p a r e d introspection to an 'inner eye', comparabie to o t h e r sense-organs. 4 5 (3) T h e d o m a i n of private experience is accessible 3. Misconceptions about privileged ,. , , , , . , , , . ,. A , r. , . directly only to the subiect: others have only indirect access and epistemic privacy ' ' , , J _ _ -. , , , . access. So, ror example, J. R . Searle held that M y present state of consciousness is a feature of m y brain, b u t its conscious aspects are accessible to m e in a w a y that they are n o t accessible to y o u . A n d y o u r present state of consciousness is a feature of y o u r brain and its conscious aspects are accessible to y o u in a w a y that they are n o t accessible to m e . ' 4 6 Libet t o o thinks that a ' C o n s c i o u s experience, as an awareness of some t h i n g or some event, is directly accessible only to the individual having that

2. Misconceptions about introspecr , . . . twn conceived as a form of inner perception

40

J. R . Searle, 'Consciousness', Annual Reviews, 23 (2000), p. 561. G. Frege, 'The thought', repr. in Collected Papers on Mathematics, Logic and Philosophy (Blackwell, Oxford, 1984), p. 361. 42 W.James, The Principles of Psychology (Holt, N e w York, 1890), vol. 1, p. 185. 43 B. Libet, 'The neural time-factor in perception, volition and free will', repr. in his Neurophysiology of Consciousness (Birkháuser, Boston, 1993), p. 368. 44 N . Humphrey, Consciousness Regained (Oxford University Press, Oxford, 1984), pp. 34f 45 N . Humphrey, 'The inner eye of consciousness', in Blakemore and Greenfield (eds), Mindwaves, pp. 379. It is striking and bizarre that here Humphrey holds that what the 'inner eye' of consciousness has as its field of view is the brain itself. O n such a conception, introspection should make PET and f M R I partly dispensable. 46 J. R. Searle, Minds, Brains and Science - the Í984 Reith Lectures (BBC, London, 1984), p. 25. It is noteworthy that Searle was subsequendy to abandon talk of'access' (see j . R. Searle, The Rediscovery of the Mind (MIT Press, Cambridge, MA., 1992), p. 98). 41

The Mereological Fallacy

87

experience, n o t to an external observer'. 4 7 B l a k e m o r e maintains that ' A l t h o u g h I assume that other h u m a n beings think in m u c h the same w a y as I d o , I have n o direct evidence about their conscious minds'. 4 8 A n d C r i c k concurs that 'Strictly speaking, each individual is certain only that h e himself is conscious'. So the m e n t a l d o m a i n is conceived to b e ' t h e inner', to b e contrasted w i t h the d o m a i n o f publicly observable behaviour, ' t h e o u t e r ' . It is n o t only privately owned, b u t also epistemically private; that is, only the subject k n o w s w i t h appropriate certainty w h a t obtains and occurs w i t h i n it. T h a t is because only t h e subject has direct access to it, and it is only direct access that confers the title of k n o w l e d g e . (4) Psychological predicates are names of i n n e r entities . . . , . . fobiects, states, events and processes), and their meanings , . . . , •, , c can b e grasped i n d e p e n d e n t l y of any conceptual c o n nection w i t h behavioural criteria. T h i s supposition is rarely articulated, y e t it is i m p o r t a n t , and it wittingly or unwittingly underpins the received n e u r o s d e n t i f i c c o n c e p t i o n o f the mental. 3 0 Certainly the supposition that o u r j u d g e m e n t s a b o u t o t h e r people's mental states or states of consciousness are inductive is n o t u n c o m m o n a m o n g neuroscientists. E d e l m a n contends that w e k n o w w h a t consciousness is 'for ourselves b u t can only j u d g e its existence i n others b y i n d u c t i v e inference'. 5 1 O t h e r s , such as C r i c k , h o l d that such inferences are analogical. 52 In so far as neuroscientists have given the question any t h o u g h t , there is a consensus that the relation b e t w e e n b e h a v i o u r a n d the m e n t a l state for w h i c h it is evidence is non-logical a n d n o n - c o n c e p t u a l . Further, as w e have seen, each person is assumed to have privileged access to his o w n experiences, a n d to b e able to speak of t h e m w i t h o u t recourse to his o w n behaviour. Evidently h e must k n o w w h a t t h e expressions that h e uses to signify his experiences m e a n . It appears that they must have a m e a n i n g that is independent of their associated behavioural expression, since a person does n o t apply these expressions to himself o n the g r o u n d s of his behaviour. T h e most plausible explanation of h o w names of experiences are e n d o w e d w i t h m e a n i n g t h e n appears to b e that we know what these words mean in virtue of associating them with the experiences we have, and to which we have privileged access. T h e neuroscientific c o n c e p t i o n of the m e n t a l is in effect c o m m i t t e d to this semantic c o n c e p t i o n , e v e n if most neuroscientists are u n a w a r e of this. This c o m m i t m e n t is of the first i m p o r t a n c e . If this c o n c e p t i o n w e r e correct, t h e n the t h o u g h t that psychological predicates are logically b o u n d up w i t h the behavioural criteria for their ascription in the third-person case w o u l d b e mistaken. If it w e r e mistaken, t h e n , to b e 4T Misconceptions about the , , . , ,. r meaning of psychological predicates

47

Libet, 'Neural time-factor', p. 368. Blakemore, The Mind Machine (BBC Books, London, 1988), p. 230. 49 Crick, Astonishing Hypothesis, p. 107. 50 Damasio argues that 'In the sentence "I see a car coming" the word see stands for a particular act of perceptual possession perpetrated by my organism (sic) . . . And the word "see" is there, . . . to translate the wordless piay unfolding in my mind' (Damasio, Feeling of What Happens, p. 186). Here 'see' is conceived to stand for, be the name of, an inner act. Humphrey explains that he derives the psychological concepts which he possesses from his own experience (Consciousness Regained, pp. 34£). 48

51 52

Edelman, Bright Air, Brilliant Fire, p. 111. Crick, Astonishing Hypothesis, p. 107.

88

Philosophical Problems in Neumsdence

sure, one could not support the accusation that neuroscientists commit a mereological fallacy by reference to the fact that nothing the brain can do satisfies the criteria for the ascription of psychological attributes to a being. For on this different account, the meaning of psychological expressions is given by 'direct' association with the attributes they signify, unmediated by behavioural criteria. Each of these four theses is misguided, involving far-reaching conceptual confusions: i. It is true that there are no experiences that are not someone's experiences (no pains unless someone is suffering, no beliefs without believers). But it does not follow that different people may not have the same pain, belief or experience, ii. Introspection is not a quasi-perceptual faculty at all, and it is not the source of knowledge about 'the inner', iii. Far from the subject having direct access to the inner, and others having only indirect access, the subject does not have access to anything inner at all (he has pains, not access to pains), and others, who observe the manifestations of the inner in behaviour thereby have as direct an access as there can possibly be. Hence too, one can be, and one often is, completely certain that another person is (or is not) conscious, is in pain, sees, hears, believes this or that, is angry or cheerful, iv. Finally, psychological terms are not names of inner entities in the sense in which typical physical terms can be loosely characterized as names of outer entities. Moreover, their meanings are not explained by association with 'inner' experience. We shall show why these ramifying conceptual commitments ((1) - (4) on pp. 85-7 above) are misconceived. To relieve neuroscientific reflection of these misconceptions will be a first step on the road to ridding it of its crypto-Cartesianism and also of its empiricist heritage. This in rain will make it possible to apprehend correctly what neuroscience has achieved in respect of shedding light on the empirical relations between the mental and the neural. We shall start by trying to clear away some misconceptions, regarding the commonly invoked dichotomy of what is 'inner' and what is 'outer'.

3.5 The Inner It is striking that, save in neuroscience or philosophy The conception of the psychological . , , , .. ,. . , , , „ lectures, no one says that toothache is something inner, attributes as inner and as mental , , , . ,. any more than one says that toothache is something 'mental'. Toothache is in a tooth, not in the mind — although, of course, it is not in the tooth as either a cavity is in the tooth or an infection is in the tooth. Why do we say that it is 'in' the tooth at all? Because it is the tooth that hurts (not the mind), and the sufferer points to the tooth when asked where he has a pain, nurses his cheek, and refrains from biting on the infected tooth. There is no such thing as mental toothache, and the expression 'mental toothache', far from being a pleonasm, is a nonsense. Of course, one would not feel toothache unless the nociceptor nerve terminals in the tooth pulp were excited, and this increased impulse firing was conveyed by the trigeminal nerve to the pons and then to the brain. But this does not imply that there is toothache in the brain, or that the brain feels toothache. Rather, these neural events are proximate causes or

The Mereological Fallacy

concomitants of the person's feeiing toothache. It is no coincidence that we speak of physical pain (since it has a bodily location) and contrast it with mental suffering. But mental suffering is not burning or stabbing pains in the mind - there are no such things. Rather, it is anguish or grief, feelings of humiliation or loss of self-respect, etc. Nevertheless, it is true that we are inclined to compare The metaphor of the inner and ., ,. ,. , , . . , . . t , pain with something inner and its behavioural manifestations with something 'outer'. The comparison is not silly, for, to be sure, one does not say that one has toothache on the basis of observing one's behaviour, whereas one does say that another person has toothache when one sees him clutching his swollen jaw and groaning. But, "Wittgenstein nicely remarked, 'We must get clearer about how the metaphor of revealing (outside and inside) is actually applied by us; otherwise we shall be tempted to look for an inside behind that which in our metaphor is the inside.' 53 This methodological advice is profound — but it needs clarification. Someone may have a pain arid not reveal it, may see Revealing and concealing ,. , , .. , * something and not say what he sees, think something and not voice his thoughts. In this sense, one may say that the mental is 'inner', in so far as it is what can, in some cases, be kept to oneself, concealed, and its manifestations suppressed. But if a person groans in pain, says what he sees or expresses his opinions, then he has 'revealed' what, in our metaphor of 'inner' and 'outer', is the inner. If he screams with pain as the dentist prods his tooth, one cannot say, 'Well, that is only behaviour {something "outer") — his pain is still concealed (it is "inner")'. If someone sincerely tells us what he thinks, we cannot say, 'These are only words — he has kept his thoughts to himself. And if he shows us what he sees, we too can see what he sees — even though we do not look into anything. (What he sees is not in his brain, and if we were to scan his brain by the most sophisticated PET, we still would not see what he sees.) Someone may feel ashamed of what he said or did and conceal his shame, but if he blushes with shame, then his shame is revealed, not concealed, by his blush. The 'inner' does not stand behind the 'outer', concealed in the brain or mind — but sometimes one may keep one's thoughts and feelings to oneself, and sometimes one may positively conceal them — suppress their natural manifestations, pretend or lie. But it should be remembered that one may conceal one's thoughts by writing one's diary in code - and there is nothing 'inner' about that. So, it is misconceived to claim, as Damasio does, that 'the Observing the inner r , - • > , , , J Í - I • i term jeeling should be reserved tor the private, mental experience of an emotion . . . this means that you cannot observe a feeling in someone else, although you can observe a feeling in yourself when, as a conscious being you perceive your own emotional states'.54 To feel angry or ashamed, to feel compassion or pity, to feel jealous or envious, is not to perceive or observe anything. And far from it being impossible to observe a feeling in another person, it is perfectly common and familiar.

L. "Wittgenstein, 'Notes for lectures on "private experience" and "sense data'", ed. R. Rhees, Philosophical Review, 11 (1968), p. 280. Damasio, Feeling of What Happens, p. 42.

90

Philosophical Problems in Neuroscience

When someone, in circumstances of danger, blanches with fear, cries out in terror, trembles and draws back from the clanger - his fear is manifest. Of course, to observe the manifestations of another person's fear is not to feel the same fear as he (although that too is possible, if we are just as frightened as he and frightened of exactly the same thing). Rather, it is to observe that he is afraid - to see the fear written all over his face, exhibited in his demeanour and behaviour. When someone is insulted and grows red with anger, raises his voice and clenches his teeth, do we not observe that he is angry? We may not feel equally angry, or indeed angry at all, but his anger is perfectly visible, and we may quail before it. There is nothing 'inner' or unobservable about the manifest emotions of others - they are exhibited to public view. All that is true is, first, that when we observe the anger or fear of another, we do not usually feel anger or fear ourselves; second, sometimes a person may feel angry or frightened and not show it.

3.6 Introspection The above remarks on the metaphor of inner and outer Privileged access and introspection, . . . . . . „ .. • ,, rt , j , may seem trivial and irrelevant. Surely, no one is conas conceived by Humphrey, Johnson' f u s e d a b o u t thaÚ T h e s a l i e n t Laird and Weiskrantz P o m t a b o u t m e n °tions of inner and outer is surely that each of us has privileged access to the contents of our own consciousness. And is this not what scientists such as Damasio mean when they insist that each person can observe only his own feelings and not the feelings of others? For each person can introspect what is in his mind, and no one can introspect what is in the mind of another person. Indeed, some biologists currently argue that the power of scrutinizing the contents of the mind is a crucial natural development. Nicholas Humphrey holds that 'a revolutionary advance in the evolution of the mind occurred when, for certain social animals, a new set of heuristic principles was devised to cope with the pressing need to model a special section of reality — the reality comprised by the behaviour of kindred animals. The trick which nature came up with was introspection . . . [the] examination of the contents of consciousness.'"''5 (But, as we shall show, the ability to reflect on what one thinks or feels is a result of the ability to say what one thinks or feels; hence it is not 'a trick which nature came up with', but a corollary of possessing a developed language.) Cognitive scientists, such as Johnson-Laird, who try to envisage computational analogues of the workings of the mind, conceive of introspection on the model of a parallel-processing device the operating system of which 'has a model of its own operation, which it uses to guide its own processes. This "self-reflective" procedure can be applied to its own output so that the system can construct a model of its own use of such models, and so on, in a series of ever-ascending levels of representation.' 56

55

Humphrey, Consciousness Regained, p. 30. Humphrey's idea is that introspection informs a creature of the contents of its own mind, and, on this basis, it can, by analogy, construct a theory or model of the minds of other animals akin to it. 56 P. N. Johnson-Laird, The Computer and the Mind (Fontana, London, 1988), p. 362.

The Mereologkal Fallacy

91

Ajid neuro-psychologists follow Weiskrantz in explaining the strange phenomenon of blind-sight in terms of a disconnection between a monitoring system and the neural events it allegedly monitors, identifying the hypothesized monitoring system with 'awareness', which is conceived to be a form of privileged access.57 Accordingly, introspection is conceived to be the operation of a neural monitor — a neural analogue of an imagined form of inner perception. The conception of consciousness or introspection as a Introsp j j form of inner vision is misconceived. The analogy between inner vision or internal sense , , our ability to say what we perceive and our perceptual faculties, on the one hand, and our ability to say what we think, feel, want, etc. and the alleged faculty of introspection, on the other, limps. It does not warrant conceiving of introspection as akin to a perceptual faculty (an 'internal sense'), or of our ability to say what we are thinking, feeling or intending to do as the upshot of any form of perception. The perceptual faculties are faculties the exercise of which is dependent upon the state of the relevant perceptual organ, on the conditions of observation, and on the observational skills of the subject. But introspection involves no perceptual organ — one does not use one's eyes, ears or any other organ in order to be able to say what one thinks or expects, wants or intends. It involves no observation - one does not observe one's thoughts, or descry one's desires or intentions. Hence, too, it involves no observation conditions. So there is no analogue of good or poor eyesight, no 'more light' or 'having a second look from closer'. Nor are there any observational skills which might be greater or lesser, and which might be honed with practice and training. There is no more a mind's eye in anything other than a wholly metaphorical sense than there is a mind's ear, nose or tongue. And where we do invoke the metaphor of the mind's eye, we speak of seeing in not with our mind's eye (see §6.3.1). „ There is, to be sure, such a thing as introspection, but it What introspection actually is . .. , _ . r c. is not a form of inner vision or observation, borne people are more skilled at introspection than others, but not because they have a better 'inner eye'. In one sense, the introspective person is he who reflects upon himself and his character, on his emotions and motives, on his attitudes and moods. Introspection, in this sense, is a form of reflexive thought, not a form of perception. It is a route to selfknowledge and self-understanding, albeit one that is beset with the perils of self-deception. In another sense, introspection is a matter of attention to one's moods and emotions, sensations and feelings. So, one may attend to the waxing and waning of one's pains in the course of the day for medical purposes, or to the changes in one's emotions and attitudes towards someone over a period of time. But to attend to one's feelings is not to perceive one's feelings; it is rather to take note of them. I may write in my diary that the pain in the morning was not too bad, but that by noon it was becoming so severe that I had to He down. But, in order to note this, I do not perceive my pain (there is no such thing as perceiving one's own pain) — I have it, and register the fact and the description of L. Weiskrantz, 'Neuropsychology and the nature of consciousness', in Blakemore and Greenfield (eds), Mindwaves, p. 319.

92

Philosophical Problems in Neuroscience

the pain in my diary. I may note that my passion for Maisy is waning, and that I am now more attracted to Daisy - but to do so, I do not observe my emotions with an inner eye. I feel my passion for Maisy diminishing - but what that means is that, for example, I think about her less frequently, realize that I am no longer eager to see her or to be in her company, dwell less in thought upon her charms than hitherto. But I do not perceive my thinkings, realizings and dwellings. My ability to register or report my feelings and emotions involves no faculty of inner vision. That ability is not to be explained in terms of a 'monitoring system', let alone a 'neural monitoring system', but, as we shall argue below, in terms of the linguistic capacities which any normal, mature language-user possesses to a greater or lesser degree.

3.7 Privileged Access: Direct and Indirect Perhaps so, one might respond, but nevertheless, it must Misconceptions concerning privileged , , , j ^ ^ , c • -¡ > c \ , , , * surely be conceded that a person has a torm or privileged access and direct/indirect access ' . ,-, access to the contents 01 his own consciousness which no one else has. For another person to find out, come to know, whether I am in pain, what I am thinking, how I am feeling, etc., he must rely on my behaviour, on what I do and say. So he has only indirect access to my mental states. Maybe the metaphor embedded in the term 'introspection' is altogether misleading. But it surely cannot be denied that I have direct access to my own states of mind, to my thoughts and feelings, and that others do not. Whether it can legitimately be denied or not depends on how it is understood. Taken in one way, it is innocuous; taken in another, it is misguided. What is true is that the grounds justifying others in ascribing psychological attributes to me are the behavioural criteria constituted by my manifestations and expressions of sensation, perception, thought, volition, etc., whereas I avow or aver what I think or feel without any criteria at all. But it does not follow that I have direct access to anything, or that others, who judge that things are thus-and-so with me on the basis of what I do and say, have only indirect access. To have a pain, to feel cheerful or depressed, to think What 'access' might be, and why , , , , . ,_, , , , , , & . that such-and-such is the case, to want to do so-and-so, c ' the ability to give expression to the . 1 1 , , , . , r to intend to act thus, and to be able to say so is not to inner is not a matter oj accessi to .-, anything ^ape access t o a n Y t n m g - One n a s access to a library, for example, if one is permitted to use it; one has access to such-and-such a closed room or garden if one legitimately has a key to it; one has access to the President if one is permitted to see him on request; and one has access to information on the Web if one has a computer with appropriate facilities. But there is nothing comparable to these cases which can be characterized as 'access' to one's pains - what is true is that when I have a pain, I can say so immediately, without evidential grounds. When a baby is in pain, he screams. He'cannot yet say 'I have a pain' or 'My stomach aches' - not through lack of access to his pain, but because he has not yet learnt to speak. When a small child is cheerful, he bounces around merrily - but he has not yet learnt to express his good cheer with the utterance 'I'm feeling cheerful'. A person who has

The Mereohgkal Fallacy

93

mastered the use of language can make assertions. He may assert that it is going to rain, or ihat the government is going to fall, or that it is wrong to hunt foxes. For such assertions he will normally have grounds — for example, that there are rain clouds on the horizon, that the opposition has mustered enough votes to-win a motion of no confidence, that cruelty is wicked. But in some cases he may be aware that the grounds he can offer in support of his assertion are insufficient to rule out the possibility that what he asserts is, or may turn out to be, false. In other cases, he may be aware that what he asserts is essentially a matter of opinion. So he will preface his assertion with the words 'I believe' or 'I think' or 'In my opinion'. Such ability to say that one thinks, believes or opines this or that does not turn on one's having privileged access to one's thoughts, beliefs and opinions or to one's thinkings, believings and opinings. Rather, in such cases (and there are others, which complicate the tale), one has grounds for an assertion, one is aware that these grounds are less than decisive, and one qualifies one's remarks accordingly. What the person needs in order sincerely to express his belief is not access to his 'mental state of believing', 58 but awareness that his grounds are less than decisive, and fall short of warranting an unqualified assertion, and mastery of the use of the qualifying expressions 'I believe', 'I think' or 'In my opinion'. The idea that others have only indirect access to or indirect Misconceptions of indirect access' ., , , . , , . . .. , _r i c c ,, ., evidence tor tne subiect s being m pain, ieeime cheerful or and indirect evidence • _, - i ,• i> > i_ • depressed, wanting a dnnk or thinking such-and-such is equally misconceived. We can speak of'indirect evidence' or of'knowing indirectly' only where it also makes sense to speak of direct evidence and of knowing directly — for to characterize evidence or knowledge as 'indirect' is meant to draw a contrast. And the contrast between having direct evidence that A is in pain and having indirect evidence thereof is not between A's having a pain and his saying so and our observing that A has a pain and saying so. For having a pain is not a form of knowledge, nor yet a kind of evidence, and the person who groans or says that he has a pain does not say so on the basis of either evidence or observation, for to have a pain, as noted above, is not to observe anything. There is no more direct a way of knowing that a person has a pain than seeing him writhe and groan in circumstances of injury or sickness. There is no more direct a way of knowing what another sees than by his showing one what he sees, and no more direct a way of knowing what he thinks than from his sincere confession. Knowing indirectly that another person is in pain might be a matter of noticing a bottle of analgesics by his bedside together with an empty glass of water. Coming to know what another person thinks on the basis of hearsay might be taken to be a case of 'indirect knowledge' — but there is nothing indirect about a confession from the horse's mouth. It is altogether mistaken to suppose that in order to know directly what another person thinks, one would have 'to "get inside" other human beings to inspect the nature of their conscious cogitations'. 39 Among other things, believing is not a mental state. For elaboration of this point, see P. M. S. Hacker 'Malcolm and Searle on "intentional mental states'", Philosophical Investigations, 15 (1992), pp. 245-75. For discussion of the concept of a mental state, see §10.2 below. Cf. Blakemore, Mind Machine, p. 231.

94

Philosophical Problems in Neurosctence

The sense in which there is literally any such thing as 'getting inside another human being' is: examining the interior of his body and brain - for there is no such thing (save figuratively) as getting inside his mind. This is not something we cannot do; rather, there is no such thing to do. That there is no such thing is no more a limitation than it is a limitation that one cannot checkmate in draughts. So, we can 'get inside' another person's brain, but no amount of investigation of another person's neural processes by means of PET will allow us to inspect his reasoning or what he is thinking. O n the other hand, we can and often do inspect a person's reasoning and the content of his thinking. We examine the sincere expression of his thought and argument. If we want to know what Newton or Kant thought, and wish to examine his reasoning, we read his writings - and there is nothing indirect about that.60

3.8 Privacy or Subjectivity One might grant that the terminology of 'direct' and Misunderstandings of the notions of , . - , . , , - i ,• u *u -Jr n ,,. . . , , . , ,,. indirect is at best misleading. Perhaps the idiom of subjectivity, privacy , and private . . ownership'of psychological attributes a c c e s s a n d Privileged access is ill-chosen. Nevertheless, it is surely indisputable that the mental has a 'subjective mode of existence'; that 'in consequence of its subjectivity, . . . pain is not equally accessible to any observer. Its existence . . . is a first-person existence.' Every pain 'must be somebody's pain'; 'every conscious state is always someone's conscious state'. Pains, we might say, are a kind of logically private entity. N o one else can have my pain, only a similar one. Another person's pain is another pain. 'And just as I have a special relation to my conscious states, which is not like my relation to other people's conscious states, so they in turn have a special relation to their conscious states, which is not like my relation to their conscious states.'61 There is nothing indisputable about these claims. O n the contrary, they exhibit further confusion. We must try to separate the wheat from the chaff. Would one also wish to say that another person cannot

fh7JZst^SZtherliefS°r

haV£ my bcUe6

°r thouShts?

Which beHef do [ have?

* Suppose Í believe that My Love will win the 3.30. So, the belief I have is: that My Love will win the 3.30. Suppose I now tell you the evidence t have for believing what I believe, and you too come to believe that My Love will win the 3.30. So now you have the very same belief as I. This might be conceded, but, it might m It would be a further error to suppose that when a person expresses his thoughts, views or opinions, he is describing them - as if he could see something invisible to others, which he then describes for their benefit. To say what one thinks is not to describe one's thoughts, but to state or express them. A. description of one's thoughts is given by characterizing them as brilliant, confused, insightful or muddled. 61 J. R. Searle, The Rediscovery of the Mind (MIT Press, Cambridge, MA, 1992), pp. 94f. Searle does not use die idiom of 'privacy', but favours 'subjectivity'. Other philosophers and, as we have seen, some neuroscientists favour 'privacy'.

The Mereologicaí Fallacy

95

rill be objected, pains are altogether different. After all, you cari t feel my pain! Only I can frel my P a m " ^ s t a n d in an altogether special relationship to my pains. If this means merely that when you have a headache, I Having the s p ^ Q n o t u s u a Uy have one, then that is true. If it means that the criteria of identity jor pains , , , , , , , . . " two people cannot have the same headache, then it is false. If we both have a dull throbbing pain in the left temple, then we do have the same cain. - Surely not; it is just that they are exactly similar, but they are nevertheless two distinct headaches! That is incorrect. There are two distinct people, with exactly the same headache. For, after all, what headache do you have? A dull throbbing one in the temples? That is precisely the same headache as I have. But surely, your headache is in your head, and mine is in mine. H o w can they be the same headache if the headaches have a different location? To this there are two replies. First, if one is worried about the problem of distinct location, one's worries should be alleviated by the example of Siamese twins with a pain at the point ofjuncture. Here one cannot say that the pains have a different location - the twins both indicate the same location. The fact that this does not alleviate one's qualms - that is, the fact that one still wishes to say that the twins have different pains (even though they are in the same place) - shows that one's worries do not really turn on the matter of location. Second, we should note that the concept of location of sensation is not like that of location of a physical object. For two people to have a pain in the same place is neither for them successively to be in the same place and to have a pain, nor for them to be Siamese twins with a pain at the point of juncture. Rather, it is for corresponding parts of their bodies to hurt in the same manner. Identity and difference of location, then, is a red herring. Being had by a subject does not _ . , , -,, - • ^ j-o, , * . . , , , One might grant that, yet still insist that dirlerent people characterize a pain, so who the sub, , , , , , o , ject of pain is does not determine c a n n o t h a v e e x a c t 1 ^ t h e s a m e ^adache. Surely, one wants to ob ect the identity of the pain he has J ' yours ls yoms' and mine IS mine ~ h o w c o u l d they be identical? Another person's headache, as we remarked above, is another headache. This is the nub of the confusion. For being mine and being yours are not identifying characteristics of the headache — these possessive phrases characterize who has a headache, not what headache is in question. Being mine is not an identifying property of the headache from which I am suffering, which might differentiate it from your headache, any more than being the colour of my eyes is an identifying property of the colour which my eyes have. If my eyes are brown and your eyes are brown, we both have eyes of the same colour. (Two red cherries don't have different colours just because the red of the first cherry 'belongs' to the first cherry and the red of the second 'belongs' to the second.) „ . . Having a pain is not owning or possessing anything Having a pain is not owning , , . , . ,, . . . ,. . any more than having a tram to catch is owning or anything; nor is it standing m a ' relation to a pain possessing something. Having a pain is not standing in any special relation to a pain, since it is not a relation at all - any more than is having a depression, which is no more than a matter of feeling depressed. To have a pain in one's foot is a matter of one's foot hurting. The criteria of identity of a pain consist in its intensity, location and phenomenological features, and if your pain and mine tally in these respects, then we both have the same

96

Philosophical Problems in Neuroscience

pain, just as if the colour of this cushion tallies willi tl , - , , . , , , , colour of that cushion in h u e , value and c h r o m a , the . , • , , i w, • ,. the t w o cushions have the same colour. W e are incline to think otherwise, because w e unwittingly construe having a pain o n the m o d e l of luvir a penny. B u t that is mistaken, for having a penny is i n d e e d a case of ownership or p o s ^ sion, and for a person to have a p e n n y is for h i m to stand i n a relationship - namely, c o w n e r s h i p or possession - to the p e n n y . B u t for a person to have a pain is n o t for him t stand in a relation to a pain, any m o r e than for a cushion to have a red c o l o u r is for ill cushion to stand in a relation to the c o l o u r red. 6 2 So, to have a pain is n o t to have access to anything. 1 k Sensation nomináis are derived . . . . , . ., , .. t „ , , particular, it is n o t to have privileged access to somethm from sensation verbs . w h i c h o n e inalienably possesses and w h i c h is, a c c o r d i n g logically private so that n o o n e else could possess it ( b u t only s o m e t h i n g like it). T o have pain is to suffer. It is for part of one's b o d y to hurt. If A's left temple aches, ihrol intermittently, a n d is sufficiently severe to stop h i m from reading, and B's left temp aches in like m a n n e r , t h e n w e say that A a n d B have the same headache. O u r disrour; a b o u t sensations and a b o u t the sameness a n d difference of sensations is an abstraction frm talk of hurtings, achings, itchings, throbbings, etc. Sensation names are useful precise] because they allow us to abstract from A's head's aching and B's head's aching i n order t focus u p o n w h a t is c o m m o n to these t w o distinct states of affairs.63 Different people ma i n d e e d have the very same pain (not just a very similar o n e ) , for w h a t counts as t w o peopl having the same pain is that the sensation each complains of should have the s.im p h e n o m e n o l o g i c a l characteristics (e.g. t h r o b b i n g , b u r n i n g , nagging, dull), b e located i the same part of the b o d y , and b e of the same intensity. T h i s is w h a t is called 'having tl same pain', and the fact that t w o p e o p l e have t h e same pain can b e an important diagnostic m a r k of their suffering from the same disease. O f course, it is true that w h e n I have a h e a d a c h e , y o u n e e d n o t have o n e . A n d it is also true that w h e n I say I have a headache, I d o n o t d o so o n the basis of perceptual observation of m y o w n behaviour, since I d o n o t have to observe myself clutching mv head in order to b e able to say that I have a h e a d a c h e - for to have a headache is not to perceive anything. H o w e v e r , I d o n o t d o so o n t h e basis of 'introspective observation", conceived as 'inner perception', either, since there is n o such thing. I just say so - give vocal expression to m y pain. O f course, I feel a h e a d a c h e , b u t to feel a headache JUSL is to Having a pain compared with , . having a penny

62

In the jargon of logicians, the distinction between qualitative and numerical identity, \\\\\á\ applies to material objects, does not apply to properties, such as colours, or to pains. T w o nuieml objects may be qualitatively identical but numerically distinct. Neither colours nor pains admit of this distinction (although in other respects they are categorially very different). 63 Reflect on the following analogy. Imagine that colour predicates were exclusively verbs, •> distinct sets of criteria for whether someone knows something: first, criteria for whetb the mental state of knowing is present, and second, criteria consisting of the performanc manifesting knowledge. But we have only the latter criteria — we do not determi whether someone knows something by reference to any kind of mental state, but only 1 reference to his evidently learning something (e.g. by being told) and by reference to the actions of his that manifest knowledge. To acquire information is not to change from 'mental state of ignorance' to a 'mental state of knowing'. To know something to be thr and-so is ability-like, hence more akin to a power or potentiality than to a state or actualii To learn that something is so is to come to be able to do a wide range of things (to infoi Knowledge is not a state, but , i• ,• ,... bears a kinship to an ability

The Cognitive Powers

149

others, to answer certain questions, to correct others, to find, locate, identify, explain things, and so forth). To forget that something is so is not to cease to be in some state, but to cease to be able to do certain things. We ask why someone is in a given state, but how someone knows. Mental states can be interrupted and later resumed (as when one's intense concentration is interrupted by a telephone conversation and later resumed), and mental states such as intense anxiety or excitement are broken off by sleep. But one cannot interrupt someone in knowing, and one does not cease to know when one falls asleep. To ask someone 'How long have you known such-and-such?' is not like asking 'How long have you been concentrating (agitated, feeling nervous)?', but more hke 'Since when have you been in a position to . . . ?', and akin to 'Since when have you been able to . . . ?'1 . If we know that things are thus-and-so, then it is possible Ktiowe g p j £Qr u s t Q ac( . Q n t | i e kas^s Q f ( j ^ information. The infórmate act on the information known . tion that things are so may provide us with reasons, in (he context of our projects, not only for acting, but also for thinking or feeling something or other (e.g. feeling pleased or angry). What one knows is what can occur as a premiss in one's reasoning from truths to the conclusions one may draw. For language-using creatures such as ourselves, to know where, when, who, what, whether or how . . . is, among other things, to be able to answer these questions. Of course, other animals, no less than human beings, know things, although their cognitive powers are less than ours; they can act for reasons, at best, only in an attenuated sense; and they cannot answer questions, but exhibit their knowledge only in their non-linguistic behaviour. Although knowledge can be said to be ability-like, being Being able to do something does ,, , ,. . ., t , t , , . able to do something is not necessarily to know anynot entail knowing ,. , . . ., , . T , thing. Indeed, it is not necessarily even knowing how to do anything. (These distinctions, as we shall see in §5.2, have an important bearing on neuroscientists' attempts to distinguish between declarative and non-declarative memory.) To justify these claims, we must clarify the relationship between being able to do something and knowing how to do something. Then we must shed some light on the relationship between knowing how to do something and knowing that something is so.

5.1.1 Being able to and knowing how to We may distinguish between innate abilities, such as the innate and acquired abilities; one. . . . . . . . ,1-1 , ' ,... . . ability to breathe, to perceive or to move one s limbs, way and two-way abilities; active , . , , . . . . „. and ac ulred and passive powers 1 abilities, such as the ability to walk or talk. We may further distinguish between one-way abilities (or powers) and two-way abilities (or powers). All inanimate abilities are one-way powers, and may be active or passive. The ability of an acid to dissolve a metal is a one-way active power: if the conditions are appropriate, the sulphuric acid will dissolve the zinc. The acid is the active agent, but it has, as it were, no choice in the matter — it cannot refrain from dissolving the metal. The liability or susceptibility of zinc to dissolve in acid is a one-way

For further discussion of the categorial characteristics of mental states, see §10.2.

150

Human Faculties and Contemporary Neuroscience

passive power. Some of the abilities of animate creatures are one-way abilities: for example the ability to see, to hear, or to feel pain. Others are two-way abilities, which the animal can exercise at will or refrain from exercising if it so chooses: for example, the ability to walk or to talk. Acquired abilities may be gained simply through natural Abilities may be acquired through , , . .,. . , c , , . maturation (e.g. the ability oí animals to engage in sexual maturation or by learning \ , i > intercourse) or through learning (which may or may not include training or teaching). Not all successful learning results in the possession of knowledge — it may result in the possession of an ability or a skill which does not involve knowing how to do anything. So, a child must learn, and indeed is taught, to be patient, or to be silent, but these do not involve acquisition of knowledge. The successful upshot of such learning is being able to do the relevant things, but not knowing, and hence remembering, the way to do them. What then distinguishes knowing how to Ffrom merely Knowing how to compared with , . ,, ,„ „ , r •-,,-> . , . ,, being able to VI r o r us to speak 01 an animal s knowing being able to ° . ° how to V, its ability must be a two-way ability, which the animal can exercise or refrain from exercising as it pleases.2 That is not sufficient, since the ability to walk, for example, is an acquired two-way ability; but, although we have to learn to walk, the upshot of learning to walk is being able to walk, not knowing how to walk. One may lose the ability to walk through paralysis, but one cannot forget (i.e. there is no such thing as forgetting) how to walk, and one cannot later be reminded or remember how to do it. To know how to do something differs from being able to do something inasmuch as knowing how to do something is knowing the way to do it (just as knowing when or where to do something is knowing the time and place to do it). To know the way to do something includes knowing the manner, means and method (where these are appropriate). Exercise of knowledge of the way to do something is plastic, adaptive and circumstance-sensitive. To know the way to do something typically involves knowing that it is done thus, and the 'thus' may, in the case of human beings, be stated or demonstrated. One can learn how to do something by experience, trial Knowing how to do something is , I T I 1 1 1 • 1 A , . , . , and error, by being trained or taught, by being shown knounng the way to do it .' . ° ; ,,, how to do it and, with human beings, by being told how to do it. What one knows how to do is something of which it makes sense to say that one has forgotten how to do it, that one remembers, recollects, or can be reminded how to do it. It also makes sense to say that one made a mistake in trying to do something one knows how to do, that one realized that one was doing it wrongly and that one tried to correct oneself. For to know how to do something is to know the way to do it, and knowing the way to do it implies an ability to distinguish between doing it correctly and doing it incorrectly. Of course, non-human animals cannot state or describe how to do something. But an animal's knowledge of the way to do something - for example, a dog's knowing 2

A very helpful discussion of knowing how is to be found in A. R. White, The Nature of Knowledge (Rowinan and Litdefield, Totowa, NJ, 1982), pp. 14-29.

The Cognitive Powers

151

how to get home from a given point - is exhibited in the plasticity of the skill in response to obstructive circumstances, the recognition of error when it occurs, and its rectification in performance. So, one may be able to do something, although it would The , , , . , u 1,1 relationship between knowing , , • ) ! , be wrong to say that one knows how to do it, and J how to and being able to ° conversely, one may know how to do something but be unable to do it. (i) Being able to do something does not imply knowing how to do it in those cases in which the concept of knowledge (and hence too of remembering and forgetting) is simply inapplicable. It may be inapplicable because knowledge is irrelevant to the kind of ability in question (e.g. to feel the heat of the fire from three feet away - a one-way passive ability - or to walk or to be quiet — two-way abilities which involve no knowledge of a way of doing anything). It may be inapplicable because knowledge is irrelevant (because categorially inappropriate) to the type of possessor of the ability (e.g. a plant's ability to grow in the shade), (ii) Knowing how to do something does not imply being able to do it. For one may know how to do things for which one has lost the physical power or lacks the strength of will. The aged tennis coach may no longer be able to play tennis, but he surely knows how to, and one may know perfecdy well how to lose weight, but be unable to. , , W e distinguish not only between being able to do someKnowing how and knowing that: ,. , , . , , .. , , thing and knowing how to do something, but also D , , between knowing how to do something and knowing that something is so. But it is mistaken to suppose, as Gilbert Ryle (who introduced and made much of the latter distinction) did, 3 that knowing how is always and essentially different from knowing that. Knowing how and knowing that are not so much two different forms which knowledge may take, as knowledge of two different kinds of thing. To know how to do something, as suggested, is to know the way to do it, and to know the way to do something is often to know, and to be able to say or show, that it is done thus-and-so.

5.1.2 Possessing Knowledge and Containing Knowledge Knowledge may be acquired, for example, by active or . . , „ . T passive perception, or by reasoning. But it may be given one in the form of the authoritative judgement or testimony of others. Indeed, it should always be borne in mind (but is often forgotten by epistemologists and by psychologists) that much of what a human being knows is not perceptual knowledge but transmitted knowledge, learnt from the written or spoken word of others. Of course, to acquire transmitted knowledge, one must be able to perceive - that is, to see (in order to read) and hear (in order to listen to what is said) — but what is learnt is not what is perceived (one sees the words one reads, but what one learns is the information they convey). Knowledge is not orAy given, in the form of passive perception or of information imparted Knowledge acquisition

G. Ryle, Concept of Mind (Hutchinson, London, 1949), ch. 2. For criticism of Ryle's distinction, see White, Nature of Knowledge, pp. 14-18, 22-8, to which we are indebted.

152

Human Faculties and Contemporary Neurosciettce

by others, it may also be attained by endeavour - by reasoning, or by discovery or detection, which may be the upshot of seeking, searching for, experimenting or trying to find out how things are. Further, it may also be received, without endeavour and independently of being given by others, by recognizing or noticing, becoming aware or conscious of something, or by realization, on the basis of information already possessed, that things are so. As we shall see in chapter 9, the distinction between knowledge attained by endeavour or given one by the word of others, on the one hand, and knowledge received in noticing, recognizing, becoming aware or realizing, on the other, is crucial for a correct grasp of the idea of becoming and then being conscious of something. . Again, it must be emphasized, it is the human being, not Knowledge and the mereoloeical , - , - , , , ,• , , , * _ _ ., , . his brain, that knows that things are thus-and-so, knows r ,. fallacy (LeDoux, Crick, Young, ..... > how to do things, and possesses the abilities constitutive of 7 . • „, . knowing something. It makes no sense, save as a misleading figure of speech, to say, as LeDoux does, that it is 'possible for your brain to know that something is good or bad before it knows exactly what it is', 4 or for Crick to speak, as we have seen he does, of the brain's learning things about the outside world. It is a confusion to speak, as j . Z. Young does, of the brain's asking and answering questions,^ and a muddle for Semir Zeki to suppose that the acquisition of knowledge is 'a primordial function of the brain', 6 since it is not the function of the brain at all. It is misguided of Colin Blakemore to write that 'Somehow the brain knows about the properties of the retina and fills in the missing information'. 7 A person who knows where the railway station is, what time the next train is, whether it is likely to be on time, who else might be on it, etc. can answer the corresponding questions. But there is no such thing as the brain knowing when . . . , where . . . , whether . . . , etc., and there is no such thing as the brain's answering such questions. It is not the brain, but the person whose brain it is, that acquires knowledge by perception, reasoning or testimony. A concept-exercising creature that can know things may be knowledgeable or ignorant, learned or untutored, an expert or a charlatan who pretends to know. But brains cannot be said to be knowledgeable, ignorant, learned, untutored, experts or charlatans — only human beings can be such things. It is equally confused to speak, as Young does, of the The brain cannot be said to contain i • > _• - , , i j • c ,_r- i brain s containing knowledge and information, which is knowledge, as books do, or to possess . , . , , . 4. , , , . , . , , f , , . T encoded in the brain mst as knowledge can be recorded J knowledge, as human beings do , „ ° in books or computers . We may say of a book that it contains all the knowledge of a lifetime's work of a scholar, or of a filing cabinet that it contains all the available knowledge, duly card-indexed, about Julius Caesar. This means that the pages of the book or the cards in the filing cabinet have written on them expressions of a large number of known truths. In this sense, the brain contains no knowledge 4

J. LeDoux, The Emotional Brain (Phoenix, London, 1998), p. 69. J. Z. Young, Programs of the Brain (Oxford University Press, Oxford, 1978), pp. 119 and 126. 6 S. Zeki, 'Abstraction and idealism', Nature, 404 (April 2000), p. 547. 7 C. Blakemore, 'The baffled brain', in R. L. Gregory and E. H. Gombrich (eds), Illusion in Nature and Art (Duckworth, London, 1973), p. 38. s Young, programs of the Brain, p. 192. 5

The Cognitive Powers

153

whatsoever. T h e r e are n o symbols in the brain that b y their array express a single p r o p o s i tion, let alone a proposition that is k n o w n to b e true. 9 O f course, in this sense a h u m a n being contains n o k n o w l e d g e either. T o possess k n o w l e d g e is n o t to contain k n o w l e d g e . A person m a y possess, for example, a smattering of k n o w l e d g e a b o u t s e v e n t e e n t h - c e n t u r y woodcuts, b u t h e contains n o n e ; H i n d ' s history of early w o o d c u t s contains a great deal of such k n o w l e d g e , b u t has n o n e . T h e brain neither possesses n o r contains any k n o w l e d g e . Libraries, b o o k s , diaries, and i n d e x cards contain information - information that h u m a n beings can l o o k u p , learn, m e m o r i z e and add t o . Similar considerations apply, as w e saw in t h e last c h a p Brains do not contain or possess , ., , . , . r . , . ter, to the idea that the brain contains m l o r m a t i o n . A 1 great deal of information is contained i n the Encyclopaedia Britannka. In that sense, there is n o n e in t h e brain. M u c h information can b e derived from a slice t h r o u g h a tree t r u n k or from a geological specimen — a n d so t o o from P E T a n d f M R I scans of the brain's activities. B u t this is not information w h i c h the brain has. N o r is it written in the brain, let alone in the 'language of the brain', 1 0 any m o r e than dendrochronological information a b o u t the severity of winters in the 1930s is w r i t t e n in the tree t r u n k in arboreal patois. Bearing in m i n d the lessons derived from examination o f Misdescriptions of commissurotomy , , . , . „ . . , , . . , i, . . the mereological fallacy in neuroscience, it should be (Cnck, Gazzamga) . ,. , , . , , . immediately obvious that the c o m m o n neuroscientific descriptions of the results of severing the corpus callosum a n d anterior commissure are again awry. After such 'split-brain' operations, patients exhibit dramatic forms of malfunctioning. This is c o m m o n l y explained (e.g. by Crick) b y reference to t h e alleged fact that 'one half of the brain appears to b e almost totally ignorant of w h a t the o t h e r half saw'. W h e n the patient is asked to explain w h y h e m o v e d his left h a n d as h e did, 'he will invent explanations based o n w h a t his left (speaking) h e m i s p h e r e saw, n o t w h a t his right h e m i sphere k n e w ' . " As w e have seen, Gazzaniga claims that 'the left brain, observing the left hand's response, interprets that response according to a context consistent w i t h its sphere of k n o w l e d g e ' . B u t if it is senseless to ascribe k n o w l e d g e to the brain, as o p p o s e d to t h e person, it is equally senseless to ascribe k n o w l e d g e (or ignorance) to o n e h e m i s p h e r e of the brain, let alone to suppose that the other hemisphere sees things. A n d it is i n c o h e r e n t to suppose that a h e m i s p h e r e of the brain has a 'sphere of k n o w l e d g e ' . T h e forms of functional dissociation c o n s e q u e n t o n c o m m i s s u r o t o m y can readily b e described w i t h o u t transgressing the b o u n d s of sense in this way (see §14.3). 9

The fact diat a neural event is correlated with a perceived object does «oí imply that the neural event is a symbol of the object (or of anything else). There is no such thing as the brain's having, using or containing lexical symbols, i.e. signs with a rule-governed use among a community of speakers and a meaning which is given by customary explanations of meaning. And there is obviously no such thing as a brain's containing or using iconic symbols. Young, Programs of the Brain, passim. For a detailed analysis of Young's misconceived analogy between neural activity and language use, see P. M. S. Hacker, 'Languages, minds and brains', in C Blakemore and S. Greenfield (eds), Mindwaves (Blackwell, Oxford, 1987), pp. 485-505. F. Crick, The Astonishing Hypothesis (Touchstone, London, 1995), p. 170.

154

Human Faculties and Contemporary Neuroscience

5.2 M e m o r y The faculty of memory is a cognitive power of human . . . . . . , . , .. , .... beings. Again, it is at best misleading to speak, as Milner , r S mre and Kandel do ^ ' o f t h e P™&eSS t h a t h a s b e e * achieved in our understanding of 'how the nervous system learns and remembers'. 12 For it is not the nervous system that learns or remembers anything at all, but the animal. And the achievements that can be hoped for are of understanding the neural processes that make it possible for animals to remember whatever they can remember. Memory is the faculty for the retention of knowledge Memory is the faculty for retention . . „ „ . . , , . . . , r, , , , , , . , , acquired. Recollecting is the bringing to mind of knowledge r, of knowledge; what is remembered ^ , ,, , need not be of the past, but must r e t a m e d " lt 1S ^ ^ P O S S l b l e t 0 r e m e m b e r oniY what be something one previously knew o n e Previously came to know or was aware of. B u t what one or was aware of remembers need have nothing to do with the past. For, apart from facts about the past that one remembers, one also learns and remembers facts concerning the present (e.g. where one's keys are), concerning the future (e.g. when the next train leaves), as well as general facts {e.g. laws of nature) that hold at all times, and truths of mathematics or logic that are atemporal. One may distinguish among the forms which memory Factual, experiential and objectual , , r t, , „ , ... ,. . , , , can take the tollowmg three, tactual memory is hnguisticmemory distinguished ° ' ° ally expressed by sentences in which the verb remember' takes as its grammatical object a that-clause: for example, 'I remember that the Battle of Hastings was fought in 1066.' Experiential memory is expressed by sentences in which the verb 'remember' is followed by a gerund which specifies a previous perceptual experience of the person: for example, 'I remember seeing. . .' (or 'hearing', 'feeling', etc.), as well as sentences of the form 'I remember F'ing . . .' and 'I remember being V'd' (where ' I 7 ' is any verb signifying something the agent might do or undergo). Clearly, I may remember that I perceived something, did or underwent something, without remembering perceiving it, doing or undergoing it, although I cannot remember perceiving, doing or undergoing something unless I remember that I perceived, did or underwent it. So experiential memory implies factual memory, but not vice versa. Objectual memory is sometimes expressed by sentences in which the verb 'remember' is followed by a direct object signifying a perceptible thing or quality: for example, 'I remember her (her smile, the house) well', 'I remember the scent of jasmine (the colour of the wall, the taste of rasberries) clearly', where a contrast with mere factual memory is intended (I may remember that she had a sweet smile, but not be able to remember it; I may remember much about the house, but not be able to visualize it). Sometimes we use such sentences to indicate our ability to conjure up images (visual or auditory) of something previously perceived. Mnemonic imagery, in this sense, although it may be common, is not logically necessary Memory is a cognitive power of , , . r , • hutnan beings, not of their nervous system (Milner, Squire and Kandei)

12

Brenda Milner, Larry R. Squire and Eric R. Kandel, 'Cognitive neuroscience and the study of memory', Neuron, 20 (1998), p. 446.

The Cognitive Powers

155

for either factual memory or experiential memory. Moreover, having a mental image of something previously perceived is no more sufficient for remembering the object antecedently perceived than is having a photograph of it, since one must also remember what one's mental image (or photograph) is an image of. Memory of objects or qualities may involve any or all of the above forms.13 Parallel to the previously discussed distinction between Remem e g knowing how and knowing that we must distinguish between "* remembering how to do something and remembering that something is so. And just as knowing how is not always essentially distinct from knowing that, so too remembering how to do something is not always essentially distinct from remembering that something is so. For to remember how to do something is to retain one's previously acquired knowledge of the way to do it. Not to have forgotten the way to Kis to remember that one Vs thus rather than thus. In many kinds of case, this is no different from remembering that the way to Kis to do such-and-such (as is obviously the case in remembering how to open a combination lock, how to integrate, how to address the Pope, or how to spell 'Edinburgh').

5.2.1 Declarative and non-declarative memory The above classification bears on a distinction widely * I J L • • - t. i invoked by neuroscientists investigating the neural . . . _ _t _ ,. _XT , _ , underpinnings of memory. The findings of Neai Cohen and Larry Squire 'suggested a fundamental distinction in the way all of us process and store information about the world' 14 : namely, between declarative and non-declarative memory}5 Declarative memory is held to be 'what is ordinarily meant by the term memory'; it is 'propositional', it can be true or false, and it is involved 'in modelling the external world and storing representations about facts and episodes'.16 Non-declarative memory is held to underlie 'changes in skilled behaviour and the ability to respond appropriately to stimuli through practice, as the result of conditioning or habit learning'. It is held to be involved in priming, in so-called habit memory ('acquired dispositions or tendencies that are specific Neuroscientists' conception of r J , , . non-declarative memory

13

For more detailed discussion, see N. Malcolm, 'Three forms of memory', in Knowledge and Certainty (Prentice-Hall, Inc., Englewood Clifls, NJ, 1963), pp. 203-21. 14 Milner, Squire and Kandel, 'Cognitive neuroscience and the study of memory', p. 450. i:> It is striking that Milner, Squire and Kandel remark that before neuroscientists drew their distinction between declarative and non-declarative memory, 'similar ideas had been proposed by philosophers and psychologists on the basis of intuition and introspection' (ibid., p. 449). The philosopher they cite is indeed Ryle, and they refer explicidy to his distinction between knowing how and knowing that. To be sure, Ryle's distinction was not drawn on the basis of either intuition or introspection, but rather on the basis of grammar. But the neuroscientists seem unaware of the flaws in Ryle's sharp differentiation between the two. Like Ryle, diey suppose that knowing how to V is the same as being able to V. They similarly assume that knowing how is never the same as knowing that. Being unaware of Ryle's mistakes, they repeat them. 16 Milner, Squire and Kandel, 'Cognitive neuroscience and the study of memory', p. 450.

156

Human Faculties and Contemporary Neurosciencc

to a set of stimuli and that guide behaviour'), and in Pavlovian conditioning (both emotional conditioning and eyeblink reaction conditioning). Ail these different phenomena are deemed to be kinds of memory, because 'performance changes as the result of experience, which justifies the term memory 1 . i? Many forms of non-declarative memory, such as habituation, sensitization and classical conditioning, are already, it is held, well developed in invertebrates. Accordingly, the gill withdrawal reflex in Aplysia was found to be modifiable by habituation, dishabituation, sensitization, classical and operant conditioning. Similar investigations were carried out on the tail flick in crayfish, feeding in Umax, phototaxis in Hermissenda. These were held to show that 'non-declarative memory storage does not depend on specialized memory neurons or systems of neurons whose only function is to store rather than process information'. 1 " Similar research on Drosophila was held to show that they 'can remember to avoid an odour that has been paired with an electric shock', but that accumulation of cAMP 'interferes with their ability to acquire and store new information'. 19 , Valuable although this research undoubtedly is, most of Misconceptions about 'declarative . . , . , , , it is not research on memory in any sense 01 the word, memory , , ,-, • , , and the extent to which it bears on actual memory has to be demonstrated. For the conceptual framework is badly confused. We shall comment below on the misconception that we store information about the world in our brains. We shall not comment on the misconception that factual memory is involved in 'modelling the external world and storing representations about facts and episodes', save to remind neuroscientists that ordinary people do not go in for 'modelling the external world', unless they are sculptors, and that they 'store representations about facts and episodes' primarily when they stick photographs into the family album. It should, however, be noted that it is misleading to say that declarative memory is 'what is ordinarily meant by the term memory'. It would be more accurate to say that declarative memory is included in what ís ordinarily meant by the term 'memory', for in the ordinary use of 'memory' we include both remembering that and remembering how as well as remembering V'ing and remembering O (i.e. experiential and objectual memory). It is claimed that what is called 'conscious recollection' is central to declarative memory but inapplicable to non-declarative memory." But it is unclear what is meant by 'conscious recollection'. It is possible that the phrase is being used to suggest that whenever one remembers that something is so, one is aware of the past occasion on which one acquired the knowledge in question; or it may mean that whenever one remembers, one is aware that one is remembering. But it is false that whenever one remembers a piece of information previously learnt — for example, that the Batde of Hastings was in 1066 or that F — ma — one remembers the event of learning it.

17

Ibid., p. 450. Were this correct, then limping as a result of injury, or impairment of hearing as a result of over-exposure to noise, would be forms of memory. 1S ¡bid., p. 454. 19 Ibid., p. 457. 20 Ibid., p. 451.

The Cognitive Powers

157

And it is also false that whenever one remembers some fact (as when one remembers to turn off the light,21 or when, having spoken to Jack earlier, one remarks to one's wife that one is going to meet Jack tomorrow), one is aware or conscious of remembering anything. It would be equally confused to suppose that when one exercises one's non-declarative memory, no conscious recollection is ever involved. For if non-declarative memory includes remembering how to do something, then one very often remembers how to do something by calling to mind the episode in which one was taught to do it, and one may well, in exercising one's memory of how to do something, be aware of the fact that one is trying to do so. Of course, neither is necessary. But then they are not necessary for declarative factual memory either. These conceptual confusions are easily rectified. The confusions about non-declarative memory, however, A iscon p ^ a r e j e e p e r p o r j.]^ habituation, sensitization, desensitizanon-dedaratwe memory . . . . . . . . .. , „ tion, classical conditioning, etc. or invertebrates, as well as the conditioned eyeblink responses and fear reactions of mammals, are not forms of memory at all. The animal may indeed have come to respond in certain ways to certain kinds of stimuli, and its responses may change - for example, accelerate - as a result of continued exposure to a stimulus. So, one might say that the animal, as a result of experience and habituation, has learnt to react more rapidly or has acquired the ability to react more rapidly. But that does not warrant characterizing the animal as remembering anything. For nothing cognitive is involved here. Not all learning is acquisition of knowledge, for sometimes it is merely acquisition of non-cognitive abilities. In the case at hand, the animal learnt neither that something is so nor the way to do anything. An accelerated reflex or a conditioned reaction is not a form of knowledge. But memory is the retention of knowledge acquired, and remembering to I 7 is the use of knowledge retained. One can indeed condition insects and molluscs to avoid a certain stimulus, but that, by itself, does not show that any knowledge was either acquired or retained. The primitive animals in question cannot be said to have come to know that things are thus-and-so, nor can they be said to have learnt the way to do anything. It is surely misconceived to hold that Drosophila 'can remember to avoid an odor that has been paired with an electric shock', 22 for all that has been shown is that Drosphilas, as a result of conditioning, learn to avoid (i.e. acquire a disposition to avoid) an odour that has been paired with an electric shock. They can be said to have acquired a primitive one-way ability, but that is not sufficient to demonstrate any form of memory whatsoever. Indeed, even in the case of a mouse that is conditioned to fear an electric shock on hearing a tone, there is no reason to take the mouse to have acquired any knowledge. All that has been shown is that conditioning produces a regular, conditioned fear reaction. Whether the neural phenomena that have been discovered to accompany such forms of conditioned reaction also characterize cases of genuine memory is possible, but it has to

21

To remember to turn off the light is to turn off the light because one knows (has not forgotten) that one must do so. Nothing need cross one's mind when one is doing so, but if asked why one did so, one would explain that it is one's obligation, that one has to do it. 22 Ibid., p. 457.

158

Human Faculties and Contemporary Neuroscience

be shown to be so. What is clear is that these studies are Whether the neural accompaniments „ ,. n T • J i c . . . . . , not actually studies 01 memory at ail. It is argued that the of conditioned reactions also accom. ^ , , , findings m question illustrate that non-declarative memoro pany genuine memory has to be & i «viy storage does not depend on specialized neurons or syss¡lown tems of neurons whose only function is to store rather than process information'. 2i But this is doubly mistaken. For, first, no form oí memory is involved in the phenomena investigated. And secondly, there is, as we shall argue below no form of storage of information in the brain.

5.2.2 Storage, retention and memory traces Failure to clarify even the basic contours of the concept „ ., , ., r , , . ; of memory is responsible tor much further unclanty in • -r n > • • - rneuroscientmc reflection on this crucial capacity. So it is confused to suggest, for example, as does LeDoux, that 'to remember is to be conscious of some past experience'. 24 For, first, what one remembers need not be anything past — it can be present, future or timeless, although, of course, one must have learnt such facts in the past. Second, what one remembers need not be an experience at all, and is not when what is remembered is, for example, the date of the Battle of Hastings. Third, to remember the date of Hastings, who Caesar's wife was, or the way home is not to be conscious of 1066, of Calpurnia, or of the way home. Moreover, even when remembering does take the form of experiential memory, to remember F'ing — that is, a past experience — is not to be conscious of that past experience; nor is remembering feeling ill last month to be conscious of feeling ill last month. It is merely to know now that one was ill then, and to know it because one then felt ill (and not because, having forgotten the episode, one was subsequently told that one previously felt ill). just as one can perceive, fail to perceive, misperceive or Mnemonic success, failure, error „ c , „ , r.. , ., . , . . , , , suffer Irom hallucinations, so too one can remember, tail and delusion distinguished to remember, misremember or suiter from mnemonic delusions. If one comes to know that things are so, and does not forget what one leamt, then one can be said to remember that things are so. One may be altogether unable to recollect something one previously knew, and such failure of memory may be a temporary lapse or it may be permanent. If one errs correctably in one's belief concerning something one previously knew, one misremembers. But one may also labour under a mnemonic delusion regarding one's own past experiences (as the Prince Regent did in thinking that he remembered fighting at Waterloo) — in which case what one believes is so out of line that it no longer amounts to a correctable error, but to a form of derangement. Much more commonly, one may think one remembers K'ing, but come to realize that one remembers only that one Vd ~ here one confuses experiential with factual memory, typically as a result of hearing the tale of one's V'ing repeated numerous times by one's parents. LeDoux's incorrect rrsupposition , e , that memory must be of the past

Ibid., p. 454. LeDoux, Emotional Brain, p. 181.

The Cognitive Powers

159

, , . , >s knowledge retained, , , , «of knowledge stored

Memory, as we have emphasized, is knowledge retained * , (including the knowledge that one perceived, did or \ & o r underwent this or that in the past, which may take the form of factual memory, of experiential memory, and may or may not involve mnemonic imagery)- But it is a confusion to suppose, as do Squire and Kandel,25 that memory is knowledge stored, let alone stored in the brain. It is confused to claim, as Milner, Squire and Kandel do, that declarative and non-declarative memories 'are stored in different brain areas',26 for there is no such thing as storing memories in the brain. Rather, the capacity to remember various kinds of things is causally dependent on different brain areas and on synaptic modifications in these areas. The notion of storage and the associated idea of memory The origins of the ideas of memory , , . lvr * ,, „ , traces long antedate neuroscience. They began Me as D J & traces and memory storage (LeDoux) . metaphors (ol wax tablets) in Plato, and as a rudimentary speculative theory in Aristotle, who conceived of memory as the storage of an impression of a percept in the heart, functionally dependent upon the humidity of the tissues. The idea of memory as a 'storehouse of ideas' runs through the empiricist tradition of the seventeenth, eighteenth and nineteenth centuries. Indeed, this conception continues to cause confusion, the metaphor being taken to be what it is merely a metaphor for. So, for example, LeDoux recently listed an array of things one might be said to have learnt and not forgotten, and queries 'What do all of these have in common?', to which he replies, 'They are things I've learned and stored in my brain'. 27 But one may surely be sceptical about the intelligibility of storing the things he cites, such as the smell of banana pudding, the meaning of the words 'halcyon days', and the rules of dominoes, in the brain. One can store smells in bottles, write down the meanings of words in dictionaries, and codify the rules of games in documents which can then be stored - but one cannot store smells, meanings of words, or rules in a brainl Of course, what LeDoux means is that these are things that he can remember — and that is right; where he errs is in the supposition that in order to be able to remember them, he must have stored them in his brain (or anywhere else). „. . , , , , It is deeply tempting to insist that while what is stored is The temptation to think that what , . , , , , . ,, . . . . , . . . obviously not what is remembered (e.g. smells, meamngs ts stored in the mind or brain is a ' , , r representation of words, or rules of games) it is a representation of what is remembered. One is inclined to think that the knowledge antecedently acquired must be stored in one's mind or brain, in the fonn of either an image or an encoded description representing what is remembered. If it were not, it seems, one would not be able to remember what one remembers — the knowledge would be unavailable to one. The classical empiricists tended to think that what is stored is stored in the mind, and the manner in which it is stored is as a mental image or picture that wtemorv JWC J

L. R. Squire and E. R. Kandel, Memory: From Mind to Molecules (Scientific American Library, New York, 1999), pp. 211-14. Milner, Squire and Kandel, 'Cognitive neuroscience and the study of memory', p. 463. LeDoux, Emotional Brain, p. 179.

160

Human

Faculties and Contemporary

Neuroscience

represents or is a copy of the original experience. Neuroscientists think that w h a t is stored is stored in t h e brain, and the m a n n e r in w h i c h it is stored is given by a pattern of synaptic connections w i t h efficacies that lead to the excitation of certain n e u r o n s u n d e r certain conditions, w h i c h excitation represents or encodes the original experience. Gazzaniga M a n g u n a n d Ivry, for example, c o n t e n d that Encoding refers to processing information to be stored. T h e encoding stage has two separate steps: acquisition and consolidation. Acquisition registers inputs in sensory buffers and sensory analysis stages, while consolidation creates a stronger representation over time. Storage, the result of acquisition and consolidation, creates and maintains a permanent record. Finally, retrieval utilises stored information to create a conscious representation or to execute a learned behaviour like a m o t o r act. W e shall e x a m i n e these pervasive ideas in a m o m e n t . B u t simply seeing the analogy b e t w e e n the classical empiricist c o n c e p t i o n and t h e c u r r e n t neuroscientific o n e should be e n o u g h to p u t us o n o u r guard. Unsurprisingly, w i t h the development of neurophysiology, , . ., ,, . , . , the obscure idea or storing m e n t a l images in the mind, , , , , , • •, and the idea that these images are unconscious until recalled, fell from favour. In its place, the c o n c e p t i o n of brain traces b e c a m e popular. A t the e n d of the n i n e t e e n t h century James w r o t e : James s conception of memory . , , . traces in the brain

J

T h e retention of ft [the previously experienced event which is n o w remembered], it wili be observed, is no mysterious storing up of an 'idea' in an unconscious state. It is not a fact of the mental order at all. It is a purely physical p h e n o m e n o n , a morphological feature, the presence of these 'paths', namely, in the finest recesses of the brain's tissue. T h e recall or recollection, on the other hand, is a psycho-physical p h e n o m e n o n , with both a bodily and a mental side. T h e bodily side is the functional excitement of the tracts and paths in question; the mental side is the conscious vision of the past occurrence, and the belief that w e experienced it before. 2 ^ It should b e n o t e d that, according to James {in this passage), the m e m o r y trace ('tract' or 'path') is n o t a condition of retaining the k n o w l e d g e acquired — that is, a c o n d i t i o n of being able to d o s o m e t h i n g — it is the storage of that k n o w l e d g e ('The retention . . . is a purely physical p h e n o m e n o n , a morphological feature'). This, as w e shall see, fails to distinguish the retention of the abilities of w h i c h k n o w i n g s o m e t h i n g consists from the neural c o n ditions for possession of those abilities, and from the storage of information in inscribed or otherwise r e c o r d e d form. T h e b a c k g r o u n d presupposition o f James's reasoning is that r e m e m b e r i n g is repeating a past experience in attenuated form 'in m e m o r y ' — 'the

2S M. S. Gazzaniga, G. R. Mañgun and R. B:' Ivry, Cognitive Neuroscience: Tlie Biology of the Mind (Norton, N e w York, 1998), pp. 247f. The equivocation over the word 'representation' is noteworthy. 29 W.James, The Principles of Psychology (Holt, N e w York 1890), vol. 1, p. 655.

The Cognitive Powers

161

onscious vision of the past o c c u r r e n c e ' , as h e puts it. This is part of the empiricist legacy, rcording to w h i c h to r e m e m b e r is to revive in one's m i n d a faint copy (an 'idea') of a previous experience (an 'impression'). T h e possibility of r e p r o d u c i n g a facsimile o f a past experience thus can b e explained, James conjectures, if t h e original experience left a 'path' 'tract' - that is, a brain trace — w h i c h , if excited again, causes the recurrence of a faint copy ° f

m e

antecedent experience t o g e t h e r w i t h a belief that o n e has 'experienced it

before'. Kohler'* conception of memory tmes

T h e idea was repeated, w i t h modifications, by such G e s t a l t p s y c h o i o g l s t s a s K o f f k a a n d Kohler,

distingulshed

a n d b e c a m e a c o m m o n p l a c e a m o n g neuroscientists. T h e

basic picture, w h i c h , as w e shall see in a m o m e n t , informs neuroscientific reflection to this day, was nicely elaborated b y K o h l e r : What does recognition mean? It means that a present fact, usually a perceptual one, makes contact with a corresponding o n e in memory, a nace, a contact which gives the present perception the character oí being k n o w n or familiar. But memory contains a tremendous number of traces, ail of them representations of previous experiences which must have been established by the processes accompanying such earlier experiences. N o w , why does the present perceptual experience make contact with the right earlier experience? This is an astonishing achievement. N o b o d y seems to doubt that the selection is brought about by the similarity of the present experience and the experience of the corresponding earlier fact. But since this earlier experience is not present at the time, w e have to assume that the trace of the earlier experience resembles the present experience, and that it is the similarity of our present experience (or the corresponding cortical process) and that trace which makes the selection possible. 30 And again: All sound theories of memory, of habit, and so forth, must contain hypotheses about memory traces as physiological facts. Such theories must also assume that the characteristics of traces are more or less akin to those of the processes by which they have been established. Otherwise, h o w could the accuracy of recall be explained, which in a great many cases is quite high. 31 Anyone holding this c o n c e p t i o n is c o m m i t t e d to the t h o u g h t that an origina] experience created a brain trace, w h i c h represents that experience. R e c o g n i t i o n is a feeling of familiarity in response to an object perceived, caused b y the excitation of a trace, w h i c h is itself caused by the neural stimulus w h i c h resembles the original cortical process. R e c o l l e c t i o n is a matter of being reminded of the antecedent experience by a current experience w h i c h resembles it in p r o d u c i n g a brain trace that corresponds (at least in part) to the trace

W. Kohler, The Task of Gestalt Psychology (Princeton University Press, Princeton, 1969), p. 122; quoted in N . Malcolm, Memory and Mind (Cornell University Press, Ithaca, NY, 1977), p. 192. W. Kohler, Gestalt Psychology (Liveright, N e w York, 1947), p. 252; quoted in Malcolm, Memory and Mind, p. 192.

162

Human

Faculties and Contemporary

Neuroscience

already laid d o w n . R e c a l l i n g s o m e t h i n g is a causal c o n s e q u e n c e of the excitation, by ; partly c o r r e s p o n d i n g neural stimulus, of the very same trace as was laid d o w n b y th< original experience. This t h o u g h t , as w e shall see, continues to inform neuroscientifn research o n m e m o r y . Squire and Kandel elaborate the jamesian conception witl Squire's and Kandel's conception of , , . - _ • , ri 1 , the sopiiistication of late t w e n t i e t h - c e n t u r y neuroscience memory storage and memory traces , , . * Conscious declarative m e m o r y , they claim, provides thi possibility of re-creating in m e m o r y a specific episode from the past'. T h e 'starting point is the set of cortical sites that w e r e engaged w h e n o n e perceived w h a t e v e r o n e perceived T h e c o n s e q u e n t m e m o r y 'uniquely depends o n t h e c o n v e r g e n c e of i n p u t from each o these distributed cortical sites into t h e medial t e m p o r a l lobe and ultimately into thh i p p o c a m p u s ' . This convergence, they claim, 'establishes a flexible representation', so tha o n e can r e m e m b e r the object perceived and the episode of perceiving it. ' T h e resultinj m e m o r i e s are stored as changes in strength at m a n y synapses w i t h i n a large ensemble o i n t e r c o n n e c t e d neurons.' F u r t h e r m o r e , 'the stored information in its specifics is determine» by t h e location of the synaptic changes', although, they admit, ' w e still k n o w relatively littL a b o u t h o w a n d w h e r e m e m o r y storage occurs'. Nevertheless, they have n o doubts tha w h a t they call 'declarative information' is stored in the brain. 3 2 This stored informatioi enables o n e to 're-create in m e m o r y ' a past episode. Ian G l y n n has recently articulated part of the presen Glynn's conception . . 1 r picture thus: Since the episodes that give rise to memories involve a variety of perceptions, it seems likely that the laying down of such memories involves nerve cells in the association areas and in secondary or higher order cortical areas concerned with the different senses. . . . It is also likely that recalling memories involves recreating something like the original pattern of activity in those same sets of cells, or at least some of them. . . . Initially then, both the hippocampal zone and the neocortical zone must act together. Eventually, w h e n consolidation is complete, the memories are stored in such a fashion that they are available without the involvement of the hippocampal zone, implying that storage is then wholly in the neocortical zone. 3 3 It should b e n o t e d that the t h o u g h t that recollecting involves re-creating a part of the original pattern of neural activity follows b o t h J a m e s a n d Kohler. A n d presumably the

32

Squire and Kandel, Memory, pp. 212f. I. Glynn, An Anatomy of Thought (Weidenfeld and Nicolson, London, 1999), p. 329. Glynn confused, however, about the subject of learning and remembering. He writes, 'what is clear is th: at the cellular and sub-cellular level machinery exists that is capable not only of simple logic, operations but also of being modified by previous experience so that its behaviour changes. It is th machinery that forms the basis of the ability of networks of nerve cells to learn and to remembei (ibid., p. 327). Of course, cellular machinery is not capable of logical operations in any literal sens - cells cannot transform propositions in accordance with rules. And it is not networks of nerve eel that remember anything, but human beings, who are able to remember what they remember i virtue of presumed changes to networks of cells in their brain. 33

The Cognitive Powers

163

motivation for this hypothesis is that it seems t o oifer t h e h o p e of an explanation of t h e Dossibility of accurate recollection, on t h e assumption that recollection is a p h e n o m e n o n 0 f re-creating in m e m o r y a specific episode from t h e past' (as Squire and Kandel p u t it). For the activation of the 'trace', like the m o v e m e n t of a stylus along the grooves of a gramophone record, is conceived to 're-create' the original experience 'in m e m o r y ' (ideas of m e m o r y b e i n g t h o u g h t to b e , as t h e empiricists held, faint copies of t h e original impressions of w h i c h they are ideas). Building o n t h e idea that recalling m e m o r i e s re-creates ,. .., , . . . . . . s o m e t h i n g like the original pattern of neural excitation, a j n vJ J i *• B e n n e t t , G i h s o n and R o b i n s o n constructed a m o d e l of the mechanism of t h e putative associative m e m o r y n e t w o r k i n t h e h i p p o c a m p u s . T h e fundamental idea is as follows: associative m e m o r y is construed as the disposition of a set of neurons (which previously fired according to a given pattern in response to a given input) to repeat the firing pattern w h e n just part of the pattern is fed i n t o t h e m . If there are x neurons in the circuit, t h e n these can b e j o i n e d w i t h connections that have such properties that a very large n u m b e r of different patterns of inputs can use different overlapping sets of these x n e u r o n s , w i t h each of these sets b e i n g m a d e to fire w h e n only a subset of the original i n p u t to the circuit is presented. T h i s m i g h t b e called 'a m e m o r i z i n g circuit'. This model is i n v o k e d to explain the neural basis for h u m a n m e m o r y . So, it is claimed, Bennett's, Gibson s ana Robinson s ' . . model of associative memory UC!

Memories are stored at the recurrent collateral synapses using a two-valued Hebbian. . . . T h e recall of a m e m o r y begins with the firing of a set of C A 3 pyramidal neurons that overlap with the m e m o r y to be recalled as well as the firing of a set of pyramidal neurons not in the memory to be recalled . . . T h e CA 3 recurrent potential network is shown to retrieve memories under specific conditions of the setting of the m e m brane potential of the pyramidal neurons by inhibitory interneurons. . . . T h e n u m b e r of memories which can be stored and retrieved without degradation is primarily a function of the number of active neurons when a m e m o r y is recalled and the degree of connectivity in the network. 3 4 T h e account provides a formal m o d e l of the brain traces envisaged by James and others. H o w e v e r , a n u m b e r of questionable ideas is i n v o l v e d in Four questionable ideas . , the received c o n c e p t i o n : i It is supposed that w h e n w e perceive s o m e t h i n g a n d r e m e m b e r w h a t w e thereby learnt, t h e n s o m e t h i n g is stored. 2 3

W h a t is stored is a m e m o r y , w h i c h represents the original perceptual experience. T h e m e m o r y is laid down in s u c h - a n d - s u c h parts of the brain, in t h e form of changes in strength at synapses. So the n e u r o n s contain a representation of the original experience.

4

Recollection involves re-creating the original pattern of activity in the relevant neurons, in particular, being r e m i n d e d o f s o m e t h i n g (associative m e m o r y ) involves having an

M. R. Bennett, W. G. Gibson and J. Robinson, 'Dynamics of the CA 3 pyramidal neuron autoassociative memory network in the hippocampus'. Philosophical Transactions of the Royal Sociely, B 334 (1994) pp. 167£

164

Human Faculties and Contemporary Neuroscience

experience which bears some similarity to the antecedent experience in which o n i acquired the information of which one is reminded by the associative experience. FQ recollecting results from stimulating the original memory trace by a neural input of; part of the original pattern of neural excitation. These four claims are disputable, and we shall raise some doubts and questions abou them. The thought that to remember is to store somethinji ., „ , . contuses retention with storage. 1 o remember is to retain . . . But although storage may sometimes imply retention retention does not imply storage. Memory, being the retention of knowledge acquired, i the retention of an ability to just the extent that knowledge itself is an ability - but it is no the storage of an ability. One may acquire and retain an ability, but that does not impi' storage. For there is no such thing as storing an ability, even though there is such a thins as retaining the neural structures that are causal conditions for the possession of an ability The supposition that if one remembers, then one must have stored a representation rest on the idea that unless there were a stored representation, the knowledge in questioi would not be available to one. H o w could one remember unless it is 'written down' ii encoded form? Encoding, as Gazzaniga, Mangun and Ivry argued (see p. 160), is process ing information into a form that will ensure that 'a permanent record' will be maintained Retrieval, which presumably means remembering, 'creates a conscious representation' b' utilizing the 'stored information'. But that is a confusion. Writing things down is indeed Neural storage of semantic . , r - . , , . r . , way oi stonng information (as long as one remember representations makes no sense how to read). Pictures do remind one of what one ha seen (as long as one remembers what the pictures are pictures of). But the idea that ii order to remember, there must be a neural record stored in the brain is incoherent. Fo even if there were such a 'record', it would not be available to a person in the sense i: which his diary or photograph album is available to him - after all, a person cannot see into his own brain, and cannot read Neuralese. Moreover, the idea that there must be a stored memory which is available to a person and is a necessary condition of his being able to remember presupposes memory (in two different ways), and cannot explain it. For were such a record available to one, one would have to remember how to read it, just as one can make use of one's diary only if one remembers how to read. Similarly, one can use one's photograph album as an aide-memoire only if one remembers what the photographs are photographs of. The idea of a store of knowledge makes sense only if the store is indeed available to one, and one can read or recognize the 'representation' - which is obviously not the case when it is supposed that the relevant information is 'stored' in the brain. The idea of neural storage of representations (in the semantic or iconic sense) is incoherent. But a neural representation in the non-semantic, non-iconic sense — that is, a causal correlate - is not a form of storing information and does not involve any encoding. (A tree does not store information about the annual rainfall in its trunk, and does not encode the rainfall in its growth bands, although we can derive such information from examination of a slice through its trunk. However, we do not derive any information Retention and storage a ... ,, , distinguished

(1)

r

The Cognitive Powers

165

whatsoever from any neural representations (in the n o n - s e m a n t i c sense) that m a y b e in o u r own brains.) If o n e perceives or learns that things are t h u s - a n d - s o , o n e ^ a s c o m e t o k n o w h o w things are. O n e m a y r e m e m b e r D J . . . what o n e has thus learnt, retain the information o n e . , acquired — that is, c o n t i n u e to possess the diffuse abilities constitutive of k n o w i n g things to be thus. T o 'retain' here simply means that o n e o n c e ]cnew and has n o t ceased to k n o w , that o n e acquired the ability, for example, t o answer the question w h e t h e r . . . or w h e r e . . . or w h e n . . . and has n o t lost it, that it b e c a m e possible for o n e to act o n t h e information that things are so, and that it is still possible for one to d o so, since o n e has n o t forgotten that things are so. N o t h i n g is implied a b o u t storage of information. 3 5 T o r e m e m b e r that p is to possess the information that p, b u t it is not to store or contain t h e information that p. O n e stores the information that p if, for example, o n e writes it d o w n , a n d stores t h e inscription in a filing cabinet or c o m p u t e r , which then contains it (but does n o t possess it). Indeed, the storage of information does not imply t h e m n e m o n i c r e t e n t i o n o f that information - there is m u c h stored i n o n e ' s filing cabinet, diary or card i n d e x w h i c h o n e has long since forgotten. B u t o n e cannot, save metaphorically, store information in one's head, a n d one's head, unlike one's diary, contains n o information. Similarly, if o n e perceives an object, place or person M , a n d o n e does n o t forget M , then one will recognize M if o n e e n c o u n t e r s it or h i m again. O n e ' s r e m e m b e r i n g M m a y therefore include, apart from facts a b o u t M , also a recognitional ability. B u t to r e m e m b e r M well, to b e able to recognize M (or a picture of M ) does n o t imply that o n e has stored anything. It implies the acquisition a n d retention of a recognitional ability. "What the neural prerequisites of this are merit investigation. To remem y thine; one cannot store information 111 '"*' , . in one s brain

(2) W h e n neuroscientists i n v o k e the n o t i o n of storage, , , , , , .. , , , their t h o u g h t is apparently that (a) w h a t is stored w h e n r r ° ' o n e ^ m e m b e r s s o m e t h i n g is a m e m o r y , (b) this stored i t e m is a representation, and (c) w h a t it represents is the original perceptual episode. T h i s is anything b u t clear. Wiiat is stored is supposed to be a .... . memory, which is a representation of an antecedent perceptual episode

A memory is not a representation

First, w e speak o f ' a m e m o r y ' a n d of h a v i n g ' m a n y happy . ,. . , ' . , „, . (or sad) m e m o r i e s of s o m e t h i n g or other. T h u s used,

the expression 'a m e m o r y ' typically signifies what is remembered w h e n o n e r e m e m b e r s that such-and-such or having such-and-such an experience. W e say such things as ' M y m e m o r y is that (things are so)', w h i c h means m u c h the same as 'As I r e m e m b e r . . . (things are so}', or 'As far as I can r e m e m b e r . . .'. W e say 'I have a d i m m e m o r y of. . . (Euclidean g e o m e t r y , Toledo, m y grandfather)', w h i c h means m u c h the same as 'I r e m e m b e r . . . b u t dimly'. So a memory is an i t e m of information (or putative information) that s u c h - a n d - s u c h or concerning this or that (or that o n e h a d s u c h - a n d - s u c h an experience), previously acquired and not forgotten. In so far as this is w h a t is m e a n t by 'a m e m o r y ' , it is evident that a m e m o r y is n o t a representation of w h a t is r e m e m b e r e d , any m o r e t h a n a belief is a

For a more detailed discussion, see Malcolm, Memory and Mind, part 2.

166

Human Faculties and Contemporary

Neuroscience

representation of w h a t is believed. B u t o n e might say that the verbal expression of w h r e m e m b e r e d is such a representation. , Secondly, a m e m o r y - that is. w h a t is remembered u One cannot store what is remem, , Á , , , , , , . > . . . , . namely that such-and-such o r having such-ana-such an exwrí rJ hexed, but only a representation of it . *r rc"~ ence - is n o t e v e n a candidate for storage. For there is n o such t h i n g as storing that such-and-such, let alone storing having an experience; at most one m i g h t store an inscription w h i c h expresses w h a t is r e m e m b e r e d or a picture which represents w h a t was experienced. B u t it w o u l d b e absurd to suppose that w h a t is allegedly stored in the brain is an English (or any other) sentence or inscription or an array oí pictures (like a p h o t o g r a p h album). 3 6 , , , Thirdly, the idea that w h a t is r e m e m b e r e d w h e n one What is remembered need not be , , . ., - - . , • , r, i J ••• r e m e m b e r s s o m e t h i n g is necessarily an original rperceor the episode of knowledge acquisition , " tual episode is mistaken. As w e have noted, w h a t we r e m e m b e r n e e d n o t b e anything past. W e h a v e all l o n g since forgotten h o w w e acquired m o s t of t h e k n o w l e d g e w e possess. I n order to b e able t o r e m e m b e r t h e m y r i a d facts w e k n o w , w e n e e d n o t , and typically d o n o t , recollect the episode o n the occasion of w h i c h w e acquired the information in question. W h e n w e learn b y reading or b y being told, for example, w h a t w e typically r e m e m b e r is what w e read or w e r e told, n o t the reading or the telling of it. B u t neuroscientists seem to suppose that the original episode of knowledge acquisition must b e 'registered' in the brain in t h e form of a representation. F o r the excitation of this representation allegedly explains three things. First, it seemingly explains the aetiology of r e m e m b e r i n g w h e n o n e is r e m i n d e d of s o m e t h i n g previously learnt. T h e current stimulus, w h i c h causes o n e to r e m e m b e r (i.e. reminds o n e of w h a t o n e then remembers), produces a neural correlate w h i c h is the same as part of the neural pattern w h i c h was laid d o w n b y the original episode. T h a t is w h y it reminds o n e of the antecede n t experience. Secondly, it explains the p h e n o m e n o n of r e m e m b e r i n g . For the stimulus excites the very same neural structure w h i c h has 'stored t h e m e m o r y ' , and that excitation 3fi

Even if remembering involved reproducing mental images of previously perceived scenes, one would still have to remember what the images were images of. Antonio Damasio holds that remembering involves not exact reproductions, but rather reconstructions, of images that are approximations to their originals. Remembering, he suggests, takes the form of 'conjurfingj up, in our mind's eye or ear, approximations of images we previously experienced'. These explicitly recalled mental images 'arise from transient synchronous activation of neural firing patterns largely in the same early sensory cortices where the firing patterns corresponding to perceptual representations once occurred' (Descartes' Error: Emotion, Reason and the Human Brain (Papermac, London, 1996), pp. 100f). To be sure, those of us who are good at producing mental images may visualize objects, scenes and events we wish to remember. But much of what we remember could not be pictured in principle (e.g. that ail Xs are F, or that no Xs are F, that if p then q, that had it been the case that p then it might have been the case that q, why such-and-such is thus-and-so, A's reason for thinking that p, and so on and so forth, through a myriad cases that spell bankruptcy for the imagist theory of memory), and anything we visualize in the course of remembering presupposes memory and cannot explain it, since we must remember what our mental image is an image of. W e shall discuss these matters in detail in the next chapter.

The Cognitive Powers

167

uses the person to have a memory experience. Thirdly, it allegedly explains why repetistrengthens or reinforces memory! For changes in strength at synapses increase with dictation.37 We shall investigate these suppositions in a moment. (3) Given that what is allegedly 'stored' or 'laid down' •j/„. idea of a neural ^ ^ h^n ^ ^ , ¿ cQuld nQt b ^ a v e r b d ^ Qr r.-niesentation is questionable . r, f' pictorial representation of an antecedent experience, it seems that the perceptions that 'give rise to memories' must be encoded in the nerve cells and synapses, and that this neural representation is what is stored. But this idea too is questionable. First, perceiving something does indeed lead to neural neid dimaperceptwn changes buthisait etherobscurewhatnu htbemeant

Zíbk°

'

°g

§

¡s

^ue by the suggestion that a perception is encoded. One can describe, in words, what one perceives and one's perceiving of it — and then encode the descriptions, assuming that one knows the transformation rules. But there is no such thing as encoding a perception. Nor is there any such thing as encoding something in the brain (at any rate, not in the ordinary sense of'encode') - for there is no such thing as a neural code. For a code is a method of encrypting a linguistic expression (or any other form of representation) according to conventional rules.38 Secondly, it is unclear what is meant by the claim that a The idea that a neural configura, , c e , ,, neural configuration may represent a memory. Suppose tion can represent a remembered tact ° . : , , , . . . the relevant memory is that one was told that the Battle is questionable . ' of Hastings was fought in 1066. What would count as a neural representation of this remembered fact? It is unclear whether, in the requisite sense of'representation', anything could count as a representation, short of an array of symbols belonging to a language. Nothing that one might find in the brain could possibly be a representation of the fact that one was told that Hastings was fought in 1066 in the sense in which the English sentence 'I was told that Hastings was fought in 1066' can be said to be such a representation.' But, of course, it may well be the case that but for certain neural configurations or strengths of synaptic connections, one would not be able to remember the date of the Battle of Hastings and would not recollect being told it. But it does not follow from that idea that what one remembers must be, as it were, written down in the brain, or that there must be some neural configuration in the brain from which one could in principle read off what is remembered. Nor can it be said that this neural configuration is a memory. It might, however, be supposed that the original perceptual episode must have caused a certain neural configuration in the brain, and the excitation of this pattern is precisely what causes one to have the experience of remembering whatever one remembers about the original episode. But even this relatively modest idea is problematic.

37

To recast die idea in the earlier jargon of engrains, the deeper the engram is 'engraved' on the brain, the more readily and vividly the memory it encodes can be invoked. 38 We have all got used to the metaphorical use of the term 'code' in the phrase 'the genetic code', tt is a metaphor that has been more damaging than illuminating.

168

Human Faculties and Contemporary

The idea that when one remembers somethfn» «n a , . . . t . •"' " given occasion, the brain must reactivate the patiom r.t 01 trace causes a mnemonic experience ° . • 1 , , , ,, ,. neural excitation that was stimulated by the antecedí..^ rests on six questionable assumptions ' •""•ci.uient perceptual experience in which one came to knou wl lt | one is now recollecting, rests upon a number of questionable assumptions: i. Currently remembering something is a (mnemonic) experience variously described as 're-creating in memory' a past episode, 'the conscious vision of a past occurrence', 01 'creating a conscious representation'. (In eighteenth-century jargon, it is having j current idea corresponding to an antecedent impression.) ii. What is remembered is a past experience or some feature of a past experience, iii. Currently remembering is triggered by a reminding experience. The remindirtt experience in certain respects resembles the past experience which is remembered. Correlative to these assumptions, current neuroscientific speculation adds a series o neuroscientific assumptions: i. Currently recollecting something is the re-excitation of a neurai firing pattern. It i< this which causally explains the person's having the mnemonic experience, ii'. The original experience now remembered laid down a neural trace (a neural structurt which, when excited, will repeat the pattern of firing generated by the original experi ence). The reactivation of the memory trace causes a mnemonic experience (tht 'conscious memory event'), which 'reproduces in memory' the original perceptua experience or some part thereof. (It is an idea or faint copy of the original impression, iii'. The current reminder triggers the brain trace, and hence the mnemonic experience by a pattern of neural input which is part of the original pattern generated by tht past perceptual experience. The similarity between the current memory experience and the original experience of which it is the memory is explained by its being caused by the reactivation of the brain trace or pattern of neural firing, which was laid down by the original perceptual experience. This reasoning, though tempting, is flawed. W e have already noted that what is remembered need not be anything past. The information acquired in the past need not be about the past, and one typically does not remember the occasion of its acquisition. All that is logically required for remembering in such cases is that one came to know something and that one still knows it (i.e. one has not forgotten the knowledge one acquired). N o w a further two points need to be stressed. First, curTo remember something need involve , , ,. , . , rt , . . rentiy remembering something need involve no reprobo reproductive representation . ductive representation . To remember that p (e.g. that the Battle of Hastings was fought in 1066, that one did such-and-such last week, that 25" - 625), to remember perceiving, doing or undergoing something, to remember M (a person, place, object or event), or to remember how to do something need not involve reproducing a 'representation', either in the form of a mental image or in the fonn of a sentence spoken aloud or in the imagination. So, for example, to remember the way home need involve no mental imagery, but only the exercise of the ability to go home without losing one's way. To remember how to drive need involve nodiing more than exercising the ability to drive. To remember what someone said need involve nothing The idea that a reactivated neural

(4)

Neuroscience

The Cognitive

Powers

169

ore than acting or reacting on the grounds of the information conveyed by the uttere j o remember perceiving, doing or undergoing something is not essentially an ability 're-create' the experience in the imagination. It is, of course, exhibited in recounting -he experience; but this is not necessary for currently remembering it. It is, as is evident rom the above examples, sufficient that one act (or react) for the reason that one previously rceived, w a s told, did or underwent such-and-such. (And here, the reason is not a cause, but a justification, which one might adduce in answer to the question of why one did what 0 ne

did.)

Secondly, remembering what we remember on a given something is no more . . . . . , -" , • , • occasion is no more an experience than is knowing what experience than is knowing . \ . ° m " ,. we know on a given occasion, inis should not be sursoinething . . . . , , . . . . . prising, since to remember that something is so just is to know something previously learnt and not forgotten. Of course, I may suddenly remember something, and this may be accompanied by various experiences {e.g. a feehng of relief, or having a mental image). But remembering something on an occasion (like knowing something) is not essentially a phenomenon; that is, it is not akin to a feeling about which one may ask 'What did it feel like?' Rather, there may be various manifestations of the fact that I remember something, none of which is the remembering. If you cancel this evening's meeting, there are indefinitely many things that I, remembering that the meeting is cancelled, may consequently do. I may go home, go to the cinema, phone any number of friends and arrange to dine together, stay on at the office until late, go to the bookshop to buy a book to read in the evening, and so forth. All these possibilities involve my remembering that the meeting is cancelled. But in none need I 're-create in memory' your cancelling this evening's meeting, or even say to myself that you have done so. All that is necessary is that part of my reason for doing what I do is that you have cancelled the meeting — and that is not an experience. If remembering something need not involve rememSo remembering need not involve , . , - i • I_-L.T I • / o experience hypothesized brain trace or pattern of neural firing in order to re-create 'in memory' a reproduction of the original episode. And if memory need involve no reproductive mnemonic experience, it is not necessary that the hypothesized brain trace be reactivated in order to produce the relevant experience. Of course, it is tempting to think that if one learnt that p (e.g. that the Battle of Hastings was fought in 1066, or that E = mc2) and one remembers this fact, then it must be 'laid down' or 'encoded' in the brain. Otherwise, how could one possibly recollect it? But, as we have seen, no sense has been given to the idea of encoding or representing factual information in the neurons and synapses of the brain. It is false that whenever one remembers something (e.g. The neuroscientific conception of the . , . . , > I I - I _ J r , . . , , t h e day of t h e w e e k , t h e w a y h o m e , o n e s last birthday, triggering of memory is confused , , , the opening bars of Beethoven s Firth) one s remembering is triggered by a current experience that causes a neural excitation similar to part of the neural excitation that was originally caused by the perceptual event in which one learnt whatever one now remembers. It is not true that whenever one's reason for doing itrmemberim ts t

170

Human Faculties and Contemporary Neuroscience

something is a fact one previously learnt, one must have been reminded of that fact bv some current experience, let alone by an experience which bears some resemblance to the past experience from which one learnt what one now recollects. Consequently, the demand that the putative brain trace be reactivated by a current experience that generates a part of the original pattern of neural excitation is altogether redundant. It was produced to meet the demands of a picture - a picture of what remembering consists in. But that picture is altogether misconceived. Neuroscientists have discovered that damage to the hippoThe neuroscientific discoveries ofthe , i-i-«~ ,_ n ,. ,. c , / . ,, campus deprives one ot the ability to recollect anything s prerequisites of retention of know, . , _ , , , , , ' . subsequently learnt or experienced lor longer than 30 t A, ledge do not support the neurosai / r o ^u entific picture of memory seconds. This certainly suggests that retention of certain neural firing patterns and synaptic connections is essential for the possibility of recollection. But it does not follow that 'memories are stored at recurrent collateral synapses', if the terms 'memory' and 'store' are being used in the normal sense. For, to repeat, there is no such thing as 'storing' what one remembers — for example, that the Battle of Hastings was fought, in Í066, or visiting Florence for the first time unless it is inscribed in symbolic form and the inscription is stored. It may be that the retention of certain synaptic connections and the creation of certain recurrent firing patterns are a necessary condition for one to be able to recall something - but that is all. The relevant synaptic connections and firing patterns cannot be said to represent 'a memory' or what is remembered. Remembering something is not a matter of retrieving something stored in the hippocampus. Nor is it having a special kind of experience — a mnemonic experience which re-creates 'in memory' some past experience. Memory is the retention of knowledge previously acquired. It is an ability that may be exercised in indefinitely many forms: for example, in saying what one remembers, alfirming that one remembers it when asked, not saying anything but thinking about what is remembered, neither saying nor thinking anything but acting on what one remembers in any of indefinitely many ways, recognizing something or someone, and so forth. It is very tempting to think that the diverse forms in which remembering something may be manifest are all due to the fact that what is remembered is recorded and stored in the brain. But that is a nonsense. What is remembered when it is remembered that such-and-such is not anything laid down in the brain, but rather something previously learnt or experienced. What neuroscientists must try to discover are the neural conditions of remembering and the neural concomitants of recollecting something. In short, neuroscientists investigating memory must distinguish between the experience of information acquisition and the information acquired, and hence between remembering the information acquired and remembering the acquiring of it. They must be careful not to slip into the error of thinking that all remembering is remembering a past experience. When they are concerned with remembering a past experience, they must not suppose that remembering need be a form of imaginative reproducing, as opposed to recounting or otherwise acting, and they must not suppose that recounting involves reading off information from a mental image. And they should avoid thinking that remembering something is a kind of experience. So, too, they must distinguish the memory - that is, what is remembered — from the expression of the memory in words, symbols or images, and also

The Cognitive Powers

111

inguish between the verbal expression of a memory and the multiple forms in which overt remembering may be manifest. The expression of a memory must be distinihed from the neural configurations, whatever they may be, which are conditions for a •on's recollecting whatever he recollects. But these configurations are not the memory; are they representations, depictions or expressions of what is remembered.

6

The Cogitative Powers

In the previous chapter we sketched the main contour lines of the general cognitive concepts of knowledge and memory. "We now turn to a group of related general concepts that similarly signify distinctive powers of human beings: belief, thought and imagination. Although not all instances of believing are cases of thinking, some are - for example, when the verb 'to believe' is used to express an opinion, what one thinks about the matter at hand. Similarly, the subject of the imagination straddles both the cogitative use of the verb 'to imagine' and its employment to signify the exercise of the power to conjure up mental images, which need have no connection with thinking. Nevertheless, the connecting thread warrants subsuming all three, at least for present purposes, under the heading of 'the cogitative powers of human beings'. Here too it is important to realize that these attributes are attributes of human beings, not of their brains. We shall not dwell on belief, which has not been the subject of much neuroscientific research. But we shall investigate thinking, imagining and mental images in some detail.

6.1 Belief The relation between knowledge A

Related to the concept of knowledge are the various ,

. , , ,.

c

i

- „•

n i-

concepts concerned with belief and conviction. .Believing falls short of knowing, for one can believe something without knowing whether things are as one believes them to be. Knowing entails that things are as they are known to be, whereas believing does not entail that things are as they are believed to be. So 'to believe', unlike 'to know', is not factive. One can know . things in detail, well or thoroughly, but one cannot believe things in any of these ways. Hence there are degrees of knowledge (in extent), but no (comparable) degrees of belief: I may know more mathematics, physics or history than you, and I may know Jack better than you do; but I cannot believe more mathematics, physics or history than you do, any more than I can believe Jack better than you-do (although I may have more faith in him and his story than you do). We may both believe that things are thus-and-so, but you cannot believe that things are so more or better than I do, and your belief cannot be greater than mine. But you may believe that things are thus with greater conviction than Í, for there

The Cogitative Powers

173

are, to be sure, degrees of conviction - that is, one may cleave to a belief more or less firmly or tenaciously. Belief, unlike knowledge, may be right or wrong, correct or incorrect, true or false. It is right, correct or true if what one believes — namely, that things are thus-and-so — is right, correct or true; that is, it is right or correct to believe, or true to say, that things are thusand-so. Belief can be firm or tentative, passionate or dogmatic, which knowledge cannot be. It is firm or tentative if one believes firmly or tentatively, and it is passionate or dogmatic if one believes passionately or dogmatically (but what one believes is neither firm or tentative, nor passionate or dogmatic). So, it is important not to confuse belief— understood as what is believed - which may be right or wrong, with belief - understood as believing — which may be passionate or dogmatic. 1 What is believed cannot be passionate or dogmatic, and believing what is believed cannot be certain, probable or possible (which what is believed can be, if it is certain, probable or possible that things are thus-and-so). Like knowing, believing is neither an act nor an activity, and it is not a feeling or a mental state either. Unlike knowing, believing is neither an ability nor akin to an ability. Like knowing, and unlike feeling sensations or perceiving tilings, one does not cease to believe whatever one believes when one falls asleep or otherwise loses consciousness. Belief is linked in various ways with doubt, certainty, . . ,, . , , . ,. , , „, conviction and being sure, hence with reeling doubttul, . , , , ,. ,. T_ certain, convinced and sure. It one believes something to be so, then one does not doubt that it is so, and one will doubt whatever one apprehends as being improbable if things are as one believes them to be. To be or feel doubtful whether something is so is to feel inclined not to believe that it is so. One can believe something to be so without being certain, sure or convinced that it is; but one cannot be certain, sure or convinced that something is so without knowing or believing that it is. Objective certainty relates to the exclusion of the possibility that things are not as one believes or knows them to be. It is certain that things are thus-and-so if the possibility of their not being so can be ruled out on the basis of the facts about the situation. Subjective certainty (feeling and being certain or sure) relates to the exclusion of doubt. Although there are no degrees of belief, there are degrees of confidence, ranging from feeling sure, certain or completely convinced to feeling uncertain or doubtful. Like the other psychological attributes, believing suchBeliefis an attribute of people, , , . ., , , , , . t ,_ , . J ,, . _ ,. and-such is ascnbable to a person, but not to his brain. not of brains (Cnck) . . .,,. L .-«»« r ^ - i Hence it is misleading or Cnck to suggest that What you see is not what is really there; it is what your brain believes is there'. There are sceptical and gullible people, but no sceptical and gullible brains. W e all know what it is The links between belief and , related attributes

One may believe truly, correcdy or righdy that p, and one may believe passionately, tentatively or firmly that p. But in the former case, the adverbs 'truly', 'correctly' and 'rightly' do not characterize the manner in which one cleaves to one's belief, whereas the adverbs 'passionately', 'tentatively' and 'firmly' do. Rather, one believes truly, correctly or rightly that p if it is true, correct or right that p. F. Crick, The Astonishing Hypothesis (Touchstone, London, 1995), p. 3 1 .

174

Human Faculties and Contemporary Neuroscience

for a person to believe or not to believe in God, to believe in the Conservative Party or in fairies, to believe a person or his story or to doubt a person's word and be sceptical about his story. But we do not know what a religious, agnostic or atheist brain might be. No sense has been given to such a form of words. There is no such thing as a brain that believes or does not believe in God, for there is no such thing as a brain's believing or not believing something. We can give this form of words a sense; we can stipulate that an agnostic brain is the brain of a person who is an agnostic — although there is no use for such a stipulation: it is altogether redundant. But we cannot stipulate that an agnostic brain is a brain that is agnostic in its beliefs about God, for brains have no beliefs - that is, there is no such thing as a brain believing or not believing something. Similarly, there are no Conservative or Labour brains, only people who believe in these parties; and the brain can neither beHeve in fairies nor believe fairy-tales, and it cannot be sceptical about them either. I may believe my friend and his story, but my brain cannot — logically cannot believe my friend or his story. _ It is noteworthy that believing that something is thusWhy belief could not be a state of , . . .. , , , , . and-so cannot - logically cannot - be a neural state of tfl€ uYHlfl

one's brain (see also §13.1 below). For if it were, and if 'I believe that such-and-such' were a report on one's believing, then it would be a report on the state of one's brain. But if it were, then since the state of one's brain is entirely independent of whether such-and-such is actually the case, it would make sense to ask the person himself whether such-and-such is the case. But if a person has said that he believes that such-and-such, he has already committed himself to the view that it is the case - which he would not have done if all he were doing were reporting on his inner state. He cannot say 'I believe that things are thus, but as to the question of whether things an thus, I have no idea (or, I have an open mind)'. But if'I believe that. . .' were a report on the state of his brain, he could intelligibly assert this nonsense. 'I believe that. . .' is typically (but not uniformly) used The use of . , , .. , , J 'I believe' to qualify an . , , , to express a magement that things are thus-and-so. assertion, to indicate the character . . . . . . . , . . , r,, , r .. . together with a rider indicating that one might be wrong & & of the grounds oj an assertion, or to i that one s take a stand on what is asserted ' S^oands for t h e assertion that things are thusand-so do not suffice to rule out the possibility that the) are not. It is not a report on how things are with one, either in one's mind or in one': brain. It is commonly akin to saying 'Such-and-such, to the best of my knowledge, is tht case', or 'Such-and-such, unless I am much mistaken, is the case* - which are not ever reports on one's mental, let alone on one's neural, state.3 Slightly differently, it is sometimes used in the same way as 'I gather that things are thus-and-so', which indicates thai I do not have first-hand knowledge. But in other contexts - for example, where knowledge is excluded by the very nature of the case — it may be used to express an opinion and so too to indicate where the speaker stands on the matter at hand. 3

For elaboration of this point, see A. W. Collins, The Nature of Mental Things (University of Notn Dame Press, Notre Dame, IN, 1987).

The Cogitative Powers

175

6.2 Thinking fhe relation between believing • thinking

Believing and thinking differ, but they do run along parallel tracks for a while. To think that something is so and to believe that something is so may be akin.4 In certain cases, there is no difference between thinking and believing, or indeed opining, that something is so. However, the concept of thinking also diverges from that of belief. For one can be engaged in thinking, but not in believing; hence one can be interrupted m thinking but not in believing. One can think through, but not believe through, a problem, and think of, as well as think up, a solution. To think of an answer to a question is not to believe of an answer, not even to believe the answer one thought of One can think* but not believe, aloud or silently, quickly or slowly, efficiently or inefficiently, fruitfully or fruitlessly. One can believe, but not think, a person, rumour or story, as indeed one can believe, but not think, in a person, a god or a cause. Thinking can be something that occurs, as when a thought The different logical categories to . , ,. , , 1 , ,. , , crosses one s mind, or something one does, as when one which thinking may belong ° thinks of a solution, as well as something one engages in, both as when one thinks through a problem for an hour and as when one engages in an activity with attention, thinking about — that is, concentrating on — what one is doing. Or it may be none of these, as when one thinks, is of the opinion that, something is thus-andso; or as when one thinks of something as something - for example, of the Epilogue to War and Peace as a stroke of genius or of Paul Klee as the most inventive artist of the twentieth century - that is how one conceives of it or of him; or as when one thinks — that is, assumes - that the bridge one is walking across is safe. There are, as indicated above, varieties of thinking. We Tlie varieties of thinking ,, . . . r J are all too prone to focus upon one particular variety to the exclusion of others, and to suppose that by studying it we are studying thinking in general. The favoured paradigm is the thinking of Le Penseur, sitting quietly, with furrowed brows, sunk in thought. As we shall see in a moment, even this paradigm conceals unexpected diversity in what, in this context, is called 'thinking'. But, what is worse, giving undue prominence to this kind of thinking obscures from sight many other things that also constitute thinking. So, first let us shift focus, and bring to mind varieties of thinking that are far removed from the cogitations of the meditator. (i) There is the thinking involved in non-meditative activities — the thinking that is constituted by attending to the task at hand. Mechanical tasks can be engaged in without

Yet they are not everywhere the same. Having been told that your rose garden is beautiful, I may ask to see it, saying 'I believe it is beautiful'. Having seen it, I would respond by telling you that I think (not that I believe) that it is beautiful. Here 'believe' would function to indicate hearsay, whereas 'think' functions to express my own opinion, gleaned at first hand. (See B. Rundle, Mind in Action (Clarendon Press, Oxford, 1997), pp. 73-80.)

176

Human Faculties and Contemporary Neurosáence

attention or concentration, indeed, when one is engager , , , j ° ° c f- , • r in such tasks — tor example, polishing the furniture o cleaning the windows — one does so without thought which does not mean that one is not thinking while polishing or cleaning, for normally one will be thinking of other things. It means that the task does not require much attentiot or care. But the more complex or delicate the task, as when one is mending a watch conducting a refined experiment, or doing a surgical operation, the more it demand concentrated attention and thought, requiring one constandy to be aware of, and to tab into account, possibilities that may obtain and difficulties that may arise. This does no mean that one is talking to oneself about such possibilities while engaged in the activity. I means that one is alert to these possible eventualities and takes precautions against them By contrast, to engage in such activities without thought or thoughtlessly is to engage in then mechanically, without due care and attention. (ii) Related to, but different from, this is the thinkin< Thinking as intelligently engaging , - • , •. s A / A 5 A ^ j IS engaging in an activity not merely with care an< in an activity , ., , . . , attention, but with cunning and ingenuity, applying one intelligence swiftly to the changing circumstances in unanticipated ways — so that one doe not merely not fall short of performing the task adequately, but rather one performs i intelligently and cleverly, as does the outstanding tennis-player or ingenious chess-mastei as well as the skilful debater in the cut and thrust of disputation. Rather differently, th brilliant actor playing Hamlet or the pianist playing the 'Hammerklavier' Sonata manifes the thoughtfuiness of their performances, not by swift and intelligent response to changin fortune and circumstance, but by the intelligence, originality and sensitivity of thei rendering. {iii) There is the thinking involved in intelligent speech. Thinking as intelligent speech „...,.-., „ , ., t ,. ,. ¿ A r This ^self is heterogeneous, l o speak without thinking may be to speak without taking into account all the factors relevant to what one >s speaking of. It is not to speak without accompanying one's speaking with an inaudible activity. By the same token, to speak with thought is to speak, taking appropriate factors into account in what one says. To be thinking hard as one speaks is not to be doing two things, speaking and thinking hard, but to be doing one thing with concentration. If, as one argues a case in public, one reasons from premisses to a certain conclusion, one is thinking, but the thinking is not something distinct from the overt reasoning manifest in one's assertion that h follows from a, b implies that either c or d, but c is incompatible with a, hence d, and so on. But speaking thoughtfully may also mean thinking before one speaks. (iv) Putting aside the kinds of thinking which involve Thinking as opining, judging, ... ,- , , 1 1 1 TO.A t- &> J A &> activities, one may think that such-and-such. I h K assuming, supposing, etc. , , . . .. , , depending on how one arrived at what one thinks and the kinds of grounds one has for it, may be a belief, an opinion, a judgement, an assumption, a supposition or presupposition, a conclusion or assessment. 'I thought it was safe (solid, robust, secure)', said ruefully after a mishap, need not mean that one reflected and came to a conclusion - it may mean that one gave the matter no thought - that is, reflection — but took it for granted, assumed or supposed — that is, thought — it Thinking as attending to a task , ,

The Cogitative Powers

177

h - thus-and-so. But it may mean that one gave the matter due attention, and came to conclusion - for example, if one says 'I knew such-and-such, so I thought (inferred) tj]:it

it was safe'.

(v) To think of something may be merely for some •flunking as assoc g object of thought to come to mind randomly, or by iVi-e.-i--' & association; but it may be a recollection, of which one n of perspective.

The Cogitative Powers

183

sensory signals'.9 A topographically arranged sensory area is not an image of anything; there are no images in the brain, and the brain does not have images. Mental ¡mages are neither necessary lt i s important to note, since it is all too often forgotten, tnat sufficient for imagining mental images are neither necessary nor sufficient for exercising the powers of the imagination. Mental images j ^ y occur or be invoked in the course of recollecting, anticipating, dreaming and daydreaming, none of which are or need be imagining. "While mental images may cross one's mind when one imagines something perceptible, they need not. One can imagine, as Tolstoy did, the Battle of Borodino without conjuring up images - what was necessary was to conjure up descriptions of what it might have been like. One can imagine, as Shakespeare did, the last words of Othello or the parting of Cassius and Brutus - what was necessary was to think up (invent) what they did or said. . . ,, Moreover, there is much that we can imagine which r , Much of what is imaginable is not ., , nv n 1 i t , . ,, could not be pictured, mentally or literally, such as what " Caesar thought on the eve of the Ides of March, why Harold Godwinson did not tarry in London to raise the fyrd, and so forth. One can imagine that all men, some men, or no men are such-and-such, but nothing in a picture or image can capture what is expressed by 'all', 'some' and 'none'. One can imagine that if jack were wiser, Jill would be happier; one can imagine what might have happened if Harald Hardrada had not invaded in the summer of 1066, or if the wind in the Channel had not changed at the very time of the Norwegian invasion. One can imagine difficulties facing one's project as well as objections to one's arguments, and so forth. A powerful imagination is not the ability to conjure up vivid mental images, but rather the ability to think of ingenious, unusual, detailed, hitherto undreamt of possibilities. And the imagination is not exercised only, or even primarily, in mere reflection, but in speech and action - in invention, creation, story-telling and problem-solving. , . , . /r , So, conjuring up mental images is not essential for the use r . The faculty of imaging (fantasia) , . , . . . . . , . , , J •f i °fr the imagination. T Indeed, the association between the is only loosely connected with the cogitative faculty of the imagination «>gitative (and creative) faculty of the imagination and the capacity to conjure up images is largely coincidental. It might be advantageous if we were to conceive of these as two distinct faculties, the first being the imagination (as conceived above), the second being fantasia.10 For to be sure, one may have a remarkably lively capacity for calling up vivid visual and auditory images, for C. Blakemore, 'Understanding images in the brain', in H. Barlow, C. Blakemore and M. WestonSmith (eds). Images and Understanding (Cambridge University Press, Cambridge, 1990), p. 282. '" See A. J. P. Kenny, The Metaphysics of Mind (Oxford University Press, Oxford, 1989), ch. 8, for a helpful discussion. Kenny has used the terms 'creative imagination' and 'fancy' where we have opted for 'cogitative imagination' and 'fantasia'. 'Cogitative imagination' links this faculty firmly to thinking, which can, but need not, be creative (when the anxious mother imagines all the dreadful things that may have happened to her absent child, she is exercising her imagination, but not being creative). 'Fantasia', not being a term with much current use, seems to us to have an advantage over 'fancy', which has a verba! use as in 'I fancied that. . .', which is equivalent to 'I imagined that. . .' (and not to 'I had an image of. . .').

184

Human

Faculties and Contemporary

Neuroscience

visualizing things and talking to oneself or rehearsing tunes to oneself 'in the imagination w i t h o u t b e i n g an imaginative person at all. A n d conversely, o n e m a y have a rich and ferti] creative imagination w i t h o u t having m u c h ability to conjure up images. T h e r e seems htt] logical c o n n e c t i o n b e t w e e n the t w o p o w e r s . I n d e e d , there seems a greater connectio b e t w e e n the faculty of m e m o r y and the faculty of imaging (fantasia) than b e t w e e n the lattt a n d the cogitative imagination. T h o s e w h o are e n d o w e d w i t h vivid mental imagery ca typically call up lively images of w h a t they previously saw, vividly picture so-and-so's fac or smile, 'relive' past experiences in their imagination. This is an exercise b o t h of m e m o r y an of fantasia, as is the ability to repeat poetry or t o rehearse a familiar piece of music to oní self in one's imagination or to r e m e m b e r so-and-so's voice w h e n h e said such-and-sucl „, . , „ T h e capacity to conjure u p m e n t a l images, in particul; / he sources of the neurosaentihc , . . , , r . . visual images, has attracted m u c h attention from neurosc D interest m mental imagery . . . entists. Part of the reason for their interest in the subja stems from the fact that they think of mental images as mysterious, publicly unobservab entities a b o u t w h i c h the cognitive scientist can m a k e discoveries, just as the physici makes discoveries about electrons or mesons. It is certainly true that o n e can mal discoveries about people and their faculty of fantasia, h e n c e about the extent to whic people's imaginative or m n e m o n i c abilities d e p e n d u p o n their p o w e r s of fantasia. B u t it misconceived to suppose that mental images are akin to theoretical entities in physics, c that studying the exercise of fantasia by means of P E T or f M R I is akin to studyir electrons in a cloud chamber. A further part of neuroscientists' interest in the subject sten from their supposition that recognition is a process of c o m p a r i n g a mental image with a perception. W h i l e it is true that machine recognition — that is, the ability of a m a c h i n e to register an object it has b e e n p r o g r a m m e d to pick o u t ('recognize') — involves matching i n p u t w i t h electronically stored images, it is a fiction that h u m a n recognition similarly involves m a t c h i n g a perception w i t h a mental image. It is claimed, o n t h e basis of extensive experiments with Posner and Raichle s hypothesis of r,,-™ , ,, J r i T , • i- • 1• ,• , . . , . . , . , P h i a n d f M R I , that visualizing s o m e t h i n g (i.e. c o n u r D w J neurai similarities between imaging ^ m g u p vlsual una and perceiving S e s o f 3t ) e v o l v e s the excitation of m u c h the same neural systems as w o u l d the correspondi n g visual experience. F u r t h e r m o r e , Michael P o s n e r a n d M a r c u s R a i c h l e c o n t e n d that ' w h e n w e construct an image, a n u m b e r of the m e n t a l operations resemble those that occur w h e n the stimulus is physically presented. In o t h e r words, there are fundamental similarities b e t w e e n perceiving an image and i m a g i n i n g o n e . M. I. Posner and M. E. Raichle, Images of Mind (Scientific American Library, New York, 1997), p. 89. It is surprising that the primary focus of neuroscientific research has been visual imaging, rather than auditory imaging. Yet talking to oneself'in the imagination' is at least as common, if not more common, that conjuring up visual images 'in the imagination'. Moreover, it is a field that is less subject to conceptual confusion than that of visual imagination. So, for example, it would be of interest to investigate the relation between the neural systems involved in, say, talking and talking to oneself in the imagination, or between those involved in listening to a piece of music, humming the same piece of music, and rehearsing the music in one's imagination, or between those involved in reading a poem, listening to that poem, reciting the poem from memory, and reciting the poem from memory in one's imagination.

The Cogitative Powers

185

Whether or not much the same neural systems are involved in the exercise of fantasia as the corresponding perceptual experience is an empirical question of some interest. But fortí this question can be resolved, the experimental scientist must be clear about the oncepts involved. In particular, he must be clear about the differences between seeing and its objects, on the one hand, and between visualizing and its objects, on the other. But many neuroscientists, active in this area, do not give adequate attention to such conceptual matters. Consequently, they fall into confusion. So, for example, Posner and Raichle assume that in permc in Positer an wppos c e p t i o n and i n fantasia alike, 'an image of the scene is ¡ng tha in p P J formed in the mind. The image formed from actual visual b alike an image is jormedm the mind . . „ . experiences is called a percept to distinguish it from an imagined image.' 12 This is mistaken. To perceive my room and the objects in it is not to form ¿n image in the mind of my room and the objects in it. To perceive, as we have seen (S4.2.3), is not to have or to form images, and what is perceived is not an image save in cases in which one perceives pictures. Moreover, to form a mental image of a scene is not to mugine an image of that scene. That is something a painter might do when he is trying to visualize the painting he intends to paint. But to form a mental image of a scene is visually to imagine that scene (not an image of it). Furthermore, the idea that there are 'fundamental similarities' between perceiving an image and imagining one is not actually what is meant. The claim is the much more general one that there are fundamental similarities between perceiving an object O and having a mental image of O, and (perhaps) also between perceiving that things are so and having an image of things being so. "What kinds of similarity do neuroscientists have in mind? One kind of similarity that is claimed is, as noted above, The alleged neural similarity between , . ., , i j c r> OÍ A , . - , , ,, that similar neural systems are involved, bo, Roger bnepard imagine and perceiving (Shepard) , . , , . , . explains that when a person conjures up a mental image ofO:

J

whatever neural activity is going on in his or her brain overlaps sufficiently with the neural activity that has previously been elicited in that same brain by the physical piesence of [O] itself, or of a picture of [O]. Thus, although it is the causal connection of the neural activity with the verbal report that informs us that someone is imagining something, it is the causal connection of that neural activity with a previously encounteied external object that defines what is being imagined.13 This claim incorporates a conceptual and an empirical component. The conceptual claim is confused, and the empirical claim cannot be correct. First, fantasia is here unwarrantedly restricted to its exercise Shepard's claim incorrectly limits . ,-, . ^. 1 . ' m memory. But one can comure up mental images oí fantasia to memory . , . innumerable objects one has never encountered. Having read Plutarch's Life of Julius Caesar, one may have mental images of Caesar and of his being 12

Ibid-, p. 88. R. Shepard, 'Postscript on understanding mental images', in Barlow, Blakemore and "WestonSmith (eds), Images and Understanding, p. 367. 13

c

186

Human Faculties and Contemporary Neuroscience

murdered in the Senate. Having been toid all about a certain person one is about to meet (or place or building one is about to visit), one may have a mental image of him and find when one encounters him, that he is (or is not) as one imagined him to be. Moreover there is much of which one may have mental images that one could not conceivably perceive, such as mental images of fictional and mythological characters (Lancelot, Pluto) or places (Camelot, Hades), as well as of fictional and mythological events. The exercise of fantasia in the context of creativity, when composing tales or music or painting mythological paintings, cannot be explained in terms of reviving antecedent impressions. Secondly, it is not a causal connection between a neural Shepard's incotrect empirical claim . . , , , . ,- , . . . , • • r , . activity and a verbal expression 01 what one is imagine: J r b concerning the entena jor the exercise .. ^ fftthat informs us that one is exercising tantasia — it is just the sincere verbal expression. What informs us whether someone has conjured up an image in his mind is what he sincerely says. Any neural concomitants are merely inductive, not criterial, evidence, and these inductive correlation* presuppose the non-inductive criterion of human avowals that determine whether a person has an image before his mind. Moreover, it is patendy not the causal connection between the neural activity and a previously encountered object that defines what is being visualized. For what the mental image I have is an image of is determined by what I sincerely say it is of, not by its causal antecedents (just as whom I meant by 'John', in m\ utterance John is at home', is determined by whom I sincerely say I meant). It is true that there is an essential connection between having visual imagery and seeing (and between having auditory imagery and hearing).14 This essential connection, however, is not that the images are derived from neural storage that was 'laid down' while seeing (or hearing). Rather, the criterion for whether someone has a visual image of something is that he say* that he has and can say how he visualizes what he imagines. For that he must have mastered the use of our ordinary vocabulary which describes visibilia, including colours and visible shapes. To have mastered such a vocabulary, he must be able to see, for without sight J person cannot, for example, master the use of colour words or describe visual appearances (save derivatively). Other similarities, suggested by Posner and Raichle, are Posner ana Raichle err in attributing . ,, , . , , ,. . . , . . , , . between the mental operations that underlie perception a similarity in the mental operations - > is •-, ,, , , , . ,. ,r , . and image formation . O n e similarity they allege is be& underlymg perception and Jantasia / / & tween the operation of scanning the letters in three-letter words and scanning a corresponding mental image. 16 But this is misconceived, since then 14

See Kenny, Metaphysics of Mind, p. 119. Posner and Raichle, Images of Mind, p. 89. 16 The experiment involved asking people to read off backwards the spelling of a word before theii eyes, and then to do the same for the mental image of a word. In the latter circumstance, people found it easy to spell 'cat' backwards, but not so easy to spell 'catapult' backwards. That they had .] mental image of the word before their mind was-of no help; and that is not surprising, since the\ could not see it (any more than they can hear their voice when they recite their favourite poem in their imagination). One can imagine a page and have a vivid image of it, but one cannot read (let alone misread) the image of a page of text. 15

The Cogitative Powers

.

187

n0 itich

thing as scanning a mental image. One cannot see — that is, there is no such thing - a mental image. I do not see the mental images I have. There is no such thing as looking at, scrutinizing or viewing a mental image, hence too, no such thing as scanning one Consequently, the conclusion drawn — namely, that 'the image is like a percept for three-letter words, but not for longer ones' — is mistaken, as is the claim that 'only a few items can be imaged at one time'. seeing

6.3.1 The logical features of mental imagery The fundamental flaws in the research and in the conTiie logical differences between , . , . /--it I J c c 11 & , f . , , elusions drawn from it stem from tailure to apprehend and fantasia have been , , , , , . „ . . .. yoerception r , . /-i.,) trie logical ditterences between perceiving an obiect and & obscured ever since Galton s psy. ,, ,havm a n , m a e o f t h a t ob ect chological questionnaire S S J > between objects and mental images of objects, and between mental images and physical images. It is noteworthy that these confusions go back to the very inception of research on mental imagery in the 1880s. Francis Galton's famous 'breakfast table' questionnaireiH queried (a) whether the brightness of a mental image is comparable to that of the actual scene; (b) whether one can 'mentally see more than three faces of a die, or more than one hemisphere of a globe at the same instant of time'; (c) whether one's mental images appear to be situated 'within the head, within the eyeball, just in front of the eyes, or at a distance corresponding to reality'; (d) whether one has ever mistaken a mental image for a reality; (e) whether one can cause one's mental images of people to sit, stand or turn slowly around; (f) whether one can see one's mental image of a person 'with enough distinctness to enable [one] to sketch it leisurely'. Most of these questions are nonsensical, and all are based on the misconceived supposition that mental images are private pictures, which one sees with the mind's eye - a supposition still rife among cognitive scientists.19 But, as we shan argue, mental images are not pictures, let alone private ones; and, pace Stephen Kosslyn and Kevin Ochsner, they are

11

Posner and Raichle, Images of Mind, pp. 89f. F. Galton, Inquiries into Human Faculty and its Development (Macmillan, London, 1883), pp. 37880. 19 e.g. 'Mental imagery is essentially a "private" or "subjective" experience, in the sense that we cannot direcdy observe other people's mental images' (J. T. E. Richardson, Imagery (Psychology Press, Hove, East Sussex, 1999), p. 9). It is true that there is no such thing as observing other people's mental images; but there is no such thing as observing one's own mental images either. The sense in which one's mental images are private is simply this: that one can have a mental image and not tell others that one has such an image or what it is an image of. If one does not tell anyone, then others will not know. i8

188

Human

Faculties and Contemporary

Neuroscience

seen in n o t with the mind's eye 2 0 - w h i c h is n o eye. T o see s o m e t h i n g in the mind's e\c is n o t to see it at all — it is visually to imagine or recollect it. Differences between perception and M , , , .. , r fantasia, between their objects, and , t , , • ,. j f between mental and physical images

W e shall spell o u t a n u m b e r of differences between , r , , . . , perception and fantasia, b e t w e e n objects perceived and

mental images of obiects, and b e t w e e n physical and meiii'il & J r ; '" images. For the concepts are m u c h more different than psychologists and cognitive scientists suppose. Fantasia is n o t a form of private perception at all - o n e can neither see n o r hear one's m e n t a l images. A m e n t a l image of an objei l is n o t related to w h a t it is a mental i m a g e of as a picture is related to w h a t it is a picture of. A n d mental images are n o t just like physical images, only m e n t a l ' — physical images and mental images are n o t co-ordinate species o f t h e genus image. (1) T h e objects of perception exist w h e t h e r o n e perObjects of perception contrasted . . , , . . , . , , . ., , . ceives t h e m or n o t , and, being m m d - m d e p e n d e n t , have with mental images , . . objective a n d d e t e r m i n a t e properties. By contrast, ment.il images d o n o t exist i n d e p e n d e n t l y of the exercise of fantasia. T h e y are, as it w e r e , made, n o t discovered. T h e i r properties can b e i n d e t e r m i n a t e , for a mental (visual) image of an object O m a y be an image of an O w i t h n o particular c o l o u r or size, even t h o u g h any C) must have s o m e colour a n d size. (A mental (visual) i m a g e is, in this respect ( b u t n o t all respects), m o r e like a drawing than like an object seen, a n d visualizing is m o r e like depicting than it is like seeing.) C o n s e q u e n t l y , Galton's question of w h e t h e r o n e can Galton's confused question n i 1 r c JL f J 1 mentally see m o r e than three faces of a die or m o r e than o n e hemisphere of a globe' at a given time is confused, and was doubtless confusing for those w h o answered his questionnaire. T o 'mentally see' — that is, to visualize or 10 conjure up a mental, visual image, is n o t to see anything. M o r e correctly phrased, the question is w h e t h e r o n e can visually imagine m o r e t h a n three faces of a given die at the same t i m e (or m o r e than one eye of a person i m a g i n e d in profile). A n d the answer is that, of course, o n e can - especially if one is familiar w i t h the paintings and drawings of Picasso. B u t t h e n , of course, o n e will n o t b e imagining (visualizing) the relevant objects as they visually appear to a perceiver. H e n c e t o o , R.. A. Finke's Principle of Spatial Equivalence, Finke's confused Principle of , . , , . (_. c , c J . . _ . , that T h e spatial a r r a n g e m e n t of the elements of a mental spatial hquwalence image corresponds to the w a y objects or their parts are arranged o n actual physical surfaces or in an actual physical space' 2 1 is at best misleading. It 20

S. M. Kosslyn and K. N . Ochsner, 'In search of occipital activation during mental imagery'. Trends in Neuroscience, 17, no. 7 (1994), pp. 290f. Visual mental imagery, they claim, 'is a quintessenually private event' — intimating that only the subject can see it. They write of the 'internal representitions that produce the experience of "seeing with the mind's eye"'. Not only is visualizing nr having a mental image of something n o t a matter of'seeing with the mind's eye', but, as we sh.il! argue, mental images are not 'representations'. 21

R. A. Finke, Principles of Mental Imagery (MIT Press, Cambridge, MA, 1989), p. 61.

The Cogitative

Powers

189

depends upon whether one has a surrealist or cubist imagination - and, of course, on what one is trying to do. If one asks one's subjects to conjure up images of what they have seen, and to imagine them as they appeared, then that is how they will be imagined. But if one asks a Picasso, a Magritte, an Escher or a Daii to imagine something ad libitum, how he imagines it will be very different from ordinary appearances, and the Principle of Spatial Equivalence will have as little validity as a corresponding principle applied to painting. (2) One uses one's sense-organs to perceive. These Ti\e contrast in respect of , , ,. . , c , 1 [ J the use organs are parts oí the body, which are under one s c of sense organs •> voluntary control in a way that affects one s powers of perception and discrimination. One may have good or bad eyes or ears, be more or less good at perceiving, depending on the condition of one's organs of perception. By contrast, there are no organs of fantasia, one cannot improve one's image of O by more careful observation of it, and one's powers of imaging are not dependent on one's organ of imaging, since there is none. Observation conditions for perception can be optimal , . , , , , or sub-optimal; one may have to approach closer, or move closer to the light, to perceive better. By contrast, there are no optimal observation conditions for the exercise of fantasia, since fantasia involves no observation. Hence, too, there is no such thing as approaching one's mental images more closely or illuminating them better in order to perceive them more clearly and distinctly, although one may imagine seeing something from closer, in good light, and conjure up an image of a well-illuminated object as it would appear from close to. One needs light in order to see an object, but one often visualizes an object better with one's eyes shut. "Whereas dazzling objects blind one temporarily, one's vivid mental image of a dazzhng object does not blind or bedazzle one, although one can imagine being bedazzled by such an object. Vie contrast in respect • °Jof observation conditions

The contrast in respect of possible error

(3)

(4) One can perceive correctly or incorrectly, make mis, , i /* ¿s takes, or overlook features otr what one perceives. One

can check whether one has done so by looking more closely in better light or asking another. By contrast, one cannot make such mistakes about one's mental images. One cannot 'overlook' features of one's mental images. They have no hidden properties which one might fail to 'discern', no occluded backsides, as it were. Rather, they are as we take them to be and as we sincerely say they are. One cannot notice or fail to notice features of one's mental images, as one can notice or fail to notice features of objects one perceives. Hence the incoherence of the explanation of a mnemonic error offered by Luria's famous mnemonist (and accepted by Luria): namely, that he imagined placing a bottle of milk in front of a white door, and then, when he subsequently imagined walking down the same street, he didn't notice the milk

Luria's mnemonist and his confused explanation of his mnemonic mistake

190

Human

Faculties and Contemporary

Neuroscience

b o t d e . 2 2 For w h a t w o u l d b e t h e criterion for the milk bottle b e i n g there ' u n n o t i c e d ' ? Xh explanation of the m n e m o n i s t ' s n o t m e n t i o n i n g a m i l k b o t d e is n o t that h e imagined on b u t did n o t notice it standing in front of a w h i t e d o o r ; it is that h e failed to include it h the scene h e revisualized. C o m p a r a b l e i n c o h e r e n c e is to b e found in the supoosi , ,. . . , ,rr r t i o n that o n e can discover things by reading off, from on? -" "m n e m o n i c image of an antecedently seen object, visitor spatial information a b o u t the object. A c c o r d i n g to thi c o n c e p t i o n , a m e n t a l image is a pictorial representation, akin to a private photograpr from w h i c h o n e can derive information a b o u t w h a t it is an image of by observation. So, fa example, Shepard claimed that in order to answer t h e question of h o w m a n y window there are in his h o u s e , h e h a d to picture the h o u s e from different sides or from withi various r o o m s and t h e n c o u n t the n u m b e r of w i n d o w s depicted i n these images. 2 3 It wa subsequently suggested b y P. R . M e u d e l l that there is a direct linear relationship betwee the time taken to answer the question and the n u m b e r of items counted. 2 4 It is uncontentiou that o n e m i g h t try to recollect h o w m a n y w i n d o w s there are in one's house b y thinkin of or i m a g i n i n g each r o o m or each elevation, a n d c o u n t i n g t h e m up as o n e proceeds. O n m a y t h e n r e m e m b e r correctly or incorrectly h o w m a n y w i n d o w s there are in o n e ' s houst W h a t is problematic, h o w e v e r , is the idea that o n e m i g h t count (and h e n c e , t o o , miscoun, t h e n u m b e r of w i n d o w s in one's mental image. Suppose there are i n fact fifteen w i n d o w s i one's h o u s e . O n e conjures u p a m e n t a l image of the different elevations, a n d conclude that there are fifteen w i n d o w s . Is it intelligible that t h e r e m i g h t have b e e n only fourteen w i n d o w s in one's mental image, w h i c h o n e m i s c o u n t e d , giving a correct answer to the question ' H o w m a n y w i n d o w s are there in y o u r h o u s e ? ' b u t an i n c o r r e c t o n e to the question ' H o w m a n y w i n d o w s are there in y o u r m e n t a l image of y o u r house?'? T h e matter is perhaps clearer in t h e case of the 'creative' The conceptual limits on counting , , t , , , . ,, . . „ r . , , . * (as o p p o s e d to the r e p r o d u c t i v e } imagination. Suppose items in one s mental image . , . i n . o n e conjures up before o n e s m i n d s eye a wholly imaginary building, perhaps an imaginary skyscraper w i t h n u m e r o u s w i n d o w s . T h e r e is n o such t h i n g as counting h o w m a n y w i n d o w s there are in the m e n t a l i m a g e . F o r it makes n o sense to try to count items w h e r e there is n o such t h i n g as miscounting, and n o such thing as rectifying a m i s c o u n t b y a recount. T h e u p p e r limit o n the determinate n u m b e r of things of Can

one derive information , . . . from one s mnemonic images by , . p r

22

A. R. Luria, The Mind of the Mnemonist (Penguin, Harmondsworth, Í968). The rnnemonist earned his living by performing extraordinary mnemonic feats. His audience would call out names of numerous objects, which the mnemonist was subsequently able to recollect in order. The heuristic device he used was to imagine walking down a familiar street in St Petersburg, and as names of objects were called out, he would imagine placing them at specific places along the street. In order to recollect the objects mentioned in the right order, he would imagine walking down the street again, and visually recollect where he had 'placed them'. 2:3 R. Shepard, 'Learning and recall as organisation and search', Journal of Verbal Learning and Verbal Behaviour, 5 (1966), pp. 201-3. 24 P. R. Meudell, 'Retrieval and representations in long-term memory', Psychonomic Science. 23 (1971), pp. 295-6.

The Cogitative Powers

191

mven kind in one's mental i m a g e is the u p p e r limit of the n u m b e r of things o n e can ecoenize at a glance to b e present (this m a y b e called 'the visual n u m b e r ' , a n d it is to b e contrasted w i t h 'the inductive n u m b e r ' 2 5 ) . O n e can visualize a battalion of soldiers m a r c h ins d o w n the C h a m p s Elysées, but there is n o such thing as counting h o w m a n y soldiers there are in the mental image of a battalion, since the n u m b e r is m u c h greater than any visual number. 2 6 O n e m a y visualize an imaginary building, and, in s o m e cases, calculate how many w i n d o w s there are in the i m a g i n e d elevation — for example, six storeys and five vrindows to a storey, so there are thirty w i n d o w s . In this w a y o n e m a y c o m e to realize that one was visualizing a h o u s e w i t h thirty w i n d o w s i n its front elevation. B u t o n e c a n n o t count thirty w i n d o w s in one's m e n t a l image of a building in order to find out h o w m a n y there are. (Similarly, o n e c a n n o t measure, b u t only imagine measuring, the objects o n e visualizes, and o n e c a n n o t w e i g h , b u t only imagine w e i g h i n g them.) (5) O n e can, in one's o w n case, mistake hallucinating One can mistake hallucinating for c . . „ , . . , . . . . , • , • for perceiving. B u t there is n o such t h i n g as mistaking r perceiving, out not fantasia ana its . ° objects for perception and its objects c o n J ™ g U P a m e n t a l i m a g e of s o m e t h i n g for perceiving that thing. Similarly, although o n e may, like M a c b e t h , mistake the hallucination of an object for the object itself, o n e c a n n o t mistake a mental image of an object for the object thus visualized. N o r can o n e mistake a m e n t a l i m a g e for an after-image or a hallucination. Gdton's confused questions

H e n c e Galton's question of w h e t h e r o n e has ever m i s . , . ,. ., . . r taken a m e n t a l image tor reality is as misguided as the

question of w h e t h e r one's obligations have ever w e i g h e d as m u c h as one's shopping bag. There is no such thing as mistaking a mental image (as opposed to an after-image or hallucination) for reality. M e n t a l images a n d w h a t they are images of d o n o t coexist in the same logical space (any m o r e t h a n heavy obligations and heavy shopping bags). Equally, Galton's question as to w h e t h e r m e n t a l images appear to b e located ' w i t h i n the head, within the eyeball, just i n front of the eyes, or at a distance c o r r e s p o n d i n g to reality' makes n o sense. A n hallucinated dagger m a y appear to M a c b e t h to b e three feet i n front of him; an after-image m a y s e e m to b e a stain o n the w h i t e wall o n e is l o o k i n g at. B u t a mental image c a n n o t appear or seem to b e located a n y w h e r e , although, of course, o n e m a y imagine Ariadne at N a x o s or at Knossos. , , , , , Mental images and physical , images contrasted

(D) W e have emphasized that m e n t a l (visual) images are ,., . . , , , . r n o t like private pictures that only the subiect can see. it , , , , ,- , , ,-^ is w o r t h elaborating further conceptual differences. As

The visual number, of course, depends upon the pattern of the array, and is subject to considerable individual differences. Of course, one may 'count sheep' in one's imagination in order to try to fall asleep, and count hundreds - but here one is not counting how many sheep there are in one's mental image of a flock of sheep, but rather how many times one imagined a sheep jumping over a stile. And when one wearily reaches 1,000, that does not mean that there are 999 sheep in one's mental image of the field on the other side of the stile.

192

Human

Faculties and Contemporary

Neuroscience

w e have noted, o n e c a n n o t see one's mental images, b u t o n e sees physical pictures physical image of an object may b e a perceptible likeness of it. B u t a mental image i u l r i b e a perceptible likeness of w h a t it is an image of, since it is n o t perceptible. O n e cannot compare one's mental image, as one can compare a physical image, witli w] it is an image o f For o n e c a n n o t juxtapose one's m e n t a l i m a g e w i t h w h a t it is an una of and l o o k at t h e m b o t h , discerning similarities and dissimilarities. Indeed, one c «ITIJ simultaneously perceive s o m e t h i n g and have an image of it as o n e perceives it to be. O n e c a n n o t copy (there is n o such t h i n g as copying) one's mental image of someihi as o n e can copy a picture of something. For, first, o n e cannot copy s o m e t h i n g thai o c a n n o t see. Secondly, a copy reproduces its original. B u t to r e p r o d u c e one's menta] n m of X , if this phrase means anything at all, w o u l d b e to imagine or visualize X again. „ , , , . Galton asked w h e t h e r , if o n e has a mental i m a r e ol dalton s confused question * person, o n e sees the m e n t a l image w i t h e n o u g h disiin t ness to enable o n e to sketch it at leisure? T h e question is confused, for it presupposes (] one's mental images are private visibilia. B u t one m i g h t ask w h e t h e r one can sketch on m e n t a l images. T o answer this curious question, o n e s h o u l d first clarify w h a t could meant by 'drawing one's mental images'. Certainly, o n e c a n n o t draw one's mental ini.u as an artist draws ( b u t does n o t copy) a person or a landscape that is before h i m . For as • have p o i n t e d out, one c a n n o t see one's mental images, and o n e c a n n o t c o m p a r e nu sketch w i t h one's m e n t a l image for verisimilitude. O n e can d r a w w h a t o n e imagines, ,i d r a w it as o n e imagines it. B u t it w o u l d b e deeply misleading in such cases to say that o is drawing one's mental images. O n e depicts what one imagines, b u t o n e does n o t iin.igi one's mental images. O n e imagines w h a t one's m e n t a l images are images of. t oiv quently, a sketch of w h a t o n e visually imagines does n o t resemble, is n o t a g o o d likeih-^ .,, one's mental image — it does n o t look like one's m e n t a l image. R a t h e r , it r e p r e s e n t I¡7ÍJ/ o n e imagined or r e m e m b e r e d and h o w o n e i m a g i n e d or r e m e m b e r e d it - just as one's w o r d s , one's descriptions of w h a t o n e imagined, d o . (7) Neuroscientists a n d cognitive scientists c h a u t r e r Wliy mental images are not . . . ,. , , ., . . , . ize m e n t a l images as internal r e p r e s e n t a t i o n s . M e n u : internal representations n i imagery is alleged to b e a form of internal representation in w h i c h information about the appearance o f physical objects, events, a n d scene-, can b e depicted and manipulated'. 2 '' B u t if pictures, maps a n d verbal descriptions are paradignis of representations, t h e n mental images are not representations of w h a t o n e im igme-.. A n y d i i n g that can b e said to be a representation of s o m e t h i n g has b o t h representanonal and non-representational properties. 2 8 N o n - r e p r e s e n t a t i o n a l properties of paintings .11 e properties of the paint (e.g. the colours of the pigments), of the canvas or panel, of ilic paint strokes (e.g. the thickness of the impasto) and paint patches (e.g. trapezoidal reel angular). Representational properties of paintings are the colours, locations and shapes ot the objects in the painting (i.e. of the painted trees, t h e painted house and the p u n t a l figures). Non-representational properties of w r i t i n g are the colours of the ink, the features Richardson, Imagery, p. 35. W e owe this point to John Hyman.

The Cogitative Powers

193

f the handwriting, and the legibility. The representational properties are the semantic roperties 0 f the words. We can apprehend the representational properties of representations only because we can perceive the non-representational ones. We could not apprehend what is said on the radio if we could not hear the sounds made. We could not read and understand what is written if we could not see the inscription. And we could not admire Rembrandt's portrait of Jan Six if we could not see the paint on the canvas. But mental images.* like thoughts, are all message and no medium. All their visual attributes are attributes of what they are images of: if one imagines red poppies in a field, the redness is of the poppies (not of any pigments), and if one imagines a dancer pirouetting rapidly, the speed j s of the imagined dancer, not of the mental image. So mental images have no nonrepresentational properties. So they are not representations. What a person says may represent how he imagines whatever he imagines. What he depicts when he draws how he imagined whatever he imagined represents what he imagined. But what he says and what he depicts do not represent what he imagined in virtue of resembling his mental imagery, any more than the verbal expression of his thought resembles his thought. What makes his sketch a good representation of what he imagined and how he imagined it is not a resemblance between the sketch and the mental image. It is not as if the sketch is a good likeness of the mental image. Rather, what makes the sketch a good representation is his sincere assertion that this is what he had in mind, this is how he imagined it. This avowal does not rest upon an 'inner glance' at his mental image. The sketch is not an 'outer picture' of an 'inner picture'. The mental image is not a representation at all. To make a representation of how one imagines something is to depict it as one imagines it, or to describe how one imagines it. It is not to conjure up an image of it. A representation represents whatever it represents in virtue of conventions of representation (semantic, cartographical, emblematic), in virtue of intended perceptible similarities between the items in the picture (the person, building or view in the picture) and the items the picture depicts (Wellington, Salisbury cathedral, the Seine). A mental image is not determined as the image it is by convention, by similarity, or by representational intention. It is not a representation of what it is a mental image of at all. „ . ., . , . (8) Because neuroscientists and cognitive scientists think Tiie vividness of mental images . . . , t , .,, jT . ., oí mental images as private pictorial representations, they r r r c t1 contrasted with the vividness of other ° suppose that having a mental image of an object can be ff-s compared with perceiving an object in the dimension of vividness or vivacity. This venerable conception was made prominent by Hume, who thought that perceiving, remembering and imagining were distinguished primarily by descending degrees of vivacity.29 We saw above that the same conception was evident in

¡

Hume, Treatise of Human Nature, I, iii, 5. Hence Thomas Reid mocked him: 'Suppose a man strikes his head smartly against the wall, this is an impression; now, he has a faculty by which he can repeat this impression with lesser force, so as not to hurt him: this, by Mr Hume's account, must be memory. He has a faculty by which he can just touch the wall with his head, so that the impression entirely looses its vivacity. This surely must be the imagination' (Essays on the Intellectual Powers, Essay III, ch. 7.)

194

Human Faculties and Contemporary Neurosaence

Galton's questionnaire, in which he asked whether 'the brightness of a mental image is comparable to that of the actual scene'. It is equally evident in Betts's questionnaire of 1909, in which he queried whether mental images were (i) perfectly clear and as vivid as the actual experience; (ii) very clear and comparable in vividness to the actual experience; (iii) moderately clear and vivid; (iv) not clear or vivid, but recognisable; (v) vague and dim; (vi) so vague and dim as to be hardly discernible.30 And it informs the current 'Vividness of Visual Imagery Questionnaire' devised b\ D. F. Marks.31 The moot question, however, is not whether mental images of objects are as vivid or less vivid than perceiving those objects. "We must first investigate what precisely is meant by ascribing vividness or vivacity to mental images, as opposed to perceiving and what is perceived. Brief reflection should make it evident that the vividness of one's main] images is more akin to the vividness of a description than to the vividness of the flaming colours of an African sunset, and that one could no more confuse the vividness of a mental image with the vividness of what it is an image of than one could confuse the vividness of a description with the vividness of what it is a description of. Some things are said to be vivid if they are full of life, vigorous and active. In this sense, people, material things and their properties can be said to be vivid ('He was a most vivid and quick-thoughted person', 'The violin is a vivid and volatile instrument', '. . . vivid and awful stenches'}. People's feelings are vivid if they are strongly felt, their utterances are vivid if they .ire strongly or warmly expressed, and their recollections are vivid if they are clear and detailed. Colour and light are vivid if they are bright, brilliant, fresh and lively, ami coloured things are accordingly said to be vivid in respect of their colour. A description, a report or a history is vivid if it presents its subject matter in a lively, clear, detailed and striking way. Consequently, a landscape is said to be vivid in the sunshine if the air is clear and the Hght is brilliant, if the different shades of green of the trees and sward are bright and distinct, if the red poppies or yellow daffodils stand out brilliandy against the green of i he grass, if the blueness of the sky and sea is intense, if the shadows are clear and sharp. In this sense, one's mental images cannot be said to be more or less vivid than what they are images of. They are not vivid and vivacious in this dimension at all. If a stage designer say* to his painter that he had imagined the redness of the backcloth to be more vivid, that does not mean that it was brighter in his imagination, any more than if an explosion was not as loud as one expected it to be, then it must have been louder in one's expectation. If the stage designer says that he imagined the light to shine more brilliantly on die

30

G. H. Betts, 'The distribution and functions of mental imagery', Contributions to Education, no. 26 (Teachers College, Columbia University, New York, 1909), pp. 20-1. 31 D. F. Marks, 'Visual imagery differences in the recall of pictures', British Journal of Psychol™)'. 64 (1973), pp. 17-24.

TTie Cogitative Powers

195

hackcloth, that does not mean that when he imagined it, it bedazzled him. One's mental images can be said to be vivid according as to whether one can give a clear, detailed and lively description of what and how one imagined (or recollected) what one imagined (or recollected), or recognize such a description as accurate if given one, or reproduce in a nainting or drawing a clear, detailed and lively representation of what one imagined (or recollected) and how one imagined (or recollected) it, or acknowledge someone else's detailed drawing (e.g. one's stage designer's) as being exactly how one imagined the scene (or set)- The brightness of one's mental image of a scene is no more comparable to the brightness of the scene than the liveliness of the description of a party is comparable to the liveliness of the party. So, too, the clarity and vividness of one's mental image of a scene are no more comparable to the clarity and vividness of the landscape bathed in brilliant evening sunshine after rain than are the clarity and vividness of one's description of the landscape. But it is true that one may be able to imagine a person more clearly than one can see him (if the light is bad and he is a long way away). What that means is that one can describe him in more detail and with greater confidence by visualizing him than by looking at the distant figure in the gloom. (9) One can often manipulate what one perceives . . . . n (depending on what it is). If it is small, one can oiten -, • , , • « -•, pick it up, turn it around, and examine it crrom all sides. One can rotate such an object quickly or slowly, at constant or variable velocity. As noted above, Galton thought that it made sense to speak of 'causing one's mental image to turn round slowly'. The moot question is whether this is intelligible, whether it makes sense to speak of manipulating mental images, and of rotating them 'in mental space' slowly or fast. Confusions about rotating mental J . , images in mental space

There is such a thing as imagining something rotating slowly {or fast), and such a tiling as having a mental image of something rotating thus. But there is no such thing as rotating a mental image of an object, any more than there is such a thing as rotating an image of an object in a painting. If one wants to have a painted picture of the object's backside, one must paint it; and if one wants to have a mental image of the backside of the object one is visualizing, one must imagine it — that is, make a mental image of it. One cannot turn one's mental image of an object around and see its backside, any more than one can rotate one of Cezanne's apples in his paintings. What one can do is imagine the object rotating. But to imagine an object rotating is not to rotate an object of any kind. Consequently, Galton's question as to whether one can Galton s confusion , , . , , r J cause one s mental images or people to sit, stand or turn slowly around is misleading. It mistakenly suggests that one might be able to turn one's mental images slowly around, whereas the only inteUigible (and trivial) question that can be asked here is whether one can visualize people turning round slowly (or quickly). One can imagine (visualize) manipulating a person (turning him around, making him sit down), but there is no such thing as manipulating one's mental image of a person. Moreover, to imagine an object rotating is not to manipulate anything — it is not to cause one's mental image to move. (One's mental image has no weight, inertia or momentum.) Galton's confusion pervades current psychological research on mental imagery. Shepard and J. Metzler presented their subjects with drawings showing perspective views of possible

196

Human

Faculties and Contemporary

Neuroscience

_, , , , , Shepard and Metzler s recognition experiments

obiects that can b e constructed b y j o i n i n g together te - , - , , n • r -\ • i i i identical cubes. Pairs oí drawings s h o w e d the atUIJ sam . object v i e w e d at different orientations, and the subjec w e r e asked to find the m a t c h i n g pairs a m o n g half a d o z e n such drawings. Shepard an Metzler found that the t i m e taken to m a t c h t w o different views of the same object w; proportional to the angle b e t w e e n the t w o views (0° — 180°), a n d that this did n o t diffi b e t w e e n objects rotated within the plane a n d those w h i c h had to b e rotated i n depth. T h suggested to t h e m that the subjects w e r e 'mentally rotating three-dimensional represent; tions' of o n e or b o t h objects at a constant rate until they h a d t h e same orientation - , w h i c h p o i n t they could easily be j u d g e d to b e t h e same or different b y simple matching. _ , , , L. A. C o o p e r and Shepard e x p e r i m e n t e d w i t h a moi Cooper s and Shepard s recognition , . *•, _ . . , , . _ . , , _ . . , , c o m pr l e x task, Subjects w e r e required to j u d g e whetht J 1 experiments; Finke s Pnnaple of Transformational Equivalence c o m m o n alpha-numeric characters w e r e presented m the n o r m a l form or as m i r r o r - i m a g e reversals. T h e subjec w e r e instructed to imagine the appropriate character in o n e o f six distinct orientations; tr character was t h e n presented either in that orientation or i n o n e of the other five, ar they h a d to j u d g e t h e m the same or different. T h e results s h o w e d that the reaction tirr again increased w i t h the angular discrepancy b e t w e e n t h e imagined orientation a n d tl orientation of the presented character. C o o p e r and Shepard inferred that the subjects rotated their m e n t a l images at constant velocity until they w e r e at the same orientation .is the presented character. T h i s suggested to C o o p e r a n d Shepard that the 'manipulation' of mental images of objects corresponds to the ways in w h i c h physical objects can be manipulated. Finke subsequendy formulated this as ' t h e Principle of Transformation.!! Equivalence': i m a g i n e d transformations and physical transformations exhibit corresponding dynamic characteristics and are g o v e r n e d by t h e same laws of m o t i o n . 3 4 T h e s e research results a n d the hypotheses that inform Posner's and Raichle's , , , , ,. . -, , . t h e m are accepted by such distinguished neuroscientists recognition experiments , ^ . , , ^ , - 1 1 • ^ 1 01 1 as Posner and R a i c h l e . Following C o o p e r a n d Shepard, they claimed that if o n e is s h o w n a letter at an angle and asked w h a t letter it is arid

32

R . N . Shepard and J. Metzler, 'Mental rotation of three-dimensional objects', Science, 171 (1971). pp. 701-3. It might be interesting to conduct the same kind of experiment with respect to matching people's faces photographed at different angles of orientation. It is probable that it takes longer to match a face seen en face with one seen in quarter-profile than with one seen in three-quarterprofile, but it is far from evident that this is to be explained in terms of rotating imagined faces. Similarly, it is plausible to suppose that it takes longer to match a young face to a very old one than to a middle-aged one, but this is not because it takes longer to age it in one's imagination. It should be noted, however, that it is a fallacy to suppose that recognition in general essentially involw< matching. To 're-cognize' is to know again, not to match something perceived with something stored or recorded. 33

L. A. Cooper and R. N . Shepard, 'Chronometric studies of the rotation of mental images', in W . G. Chase (ed.), Visual Information Processing (Academic Press, N e w York, 1973), pp. 75-176. 34 Finke, Principles of Mental Imagery, p. 93. 35 Posner and Raichle, Images of mind, p. 35.

The Cogitative Powers

197

whether it is correctly oriented or a mirror image, one's reaction time increases with the „\e of rotation in roughly linear fashion, because it takes longer to perforin the internal rotatwri than to tell which letter it is. It may weli take longer to determine whether the letter is a mirror image or not than to determine which letter it is, but not because it takes jotiger to rotate the image of the letter in one's mind, since there is no such thing as rotating a ¡nental image. The fact that the reaction time, in all these experiments, Tí;c confusions underly, , , _ . _ , _ 111 conceptual r .. . , is proportional to the angle of rotation or the figures mg the recognition experiments and .r ,r , , , . , . ° ¿ interpretation of their results visualized does not suggest that it takes longer to perform a greater rotation at constant velocity 'in mental space' than to perform a lesser rotation, since there is no such thing as rotating a mental image at constant (or variable) velocity — only such a thing as imagining an object rotating at constant (or variable) velocity. What it does suggest is that it may take longer to work out how a certain figure will appear when rotated thus than to work out how the same or another figure will appear when rotated otherwise. For one needs to exercise one's imagination — that is, one's powers to think of possibilities, to work out where this part of the figure will lie in relation to that part if the whole figure is rotated by 90°. One needs to think about the rotation of a figure, not to rotate an imaginary figure (since there is no such thing). In so thinking, one may, but need not, imagine a rotating figure. (And it is important to remember that thinking about something does not imply saying anything to oneself.) One can imagine a rotating object. But to imagine an object moving quickly does not mean that anything moved quickly in the imagination. A choreographer may say to his dancers that he imagined a particular pas de deux much faster than they danced it, but that does not mean that they danced more quickly in his imagination. If someone insists that by the phrase 'they danced more quickly in my imagination' he just means 'I imagined them dancing more quickly', we can accept that, as long as it is clear that 'they danced more quickly in my imagination' does not imply that they danced more quickly. To imagine something louder does not imply that it was louder in one's imagination, just as to have expected an explosion to be noisier than it was does not imply that it made more noise in one's expectation. The idea that it must take longer to imagine rotating a figure by 90° than to imagine rotating it by 45° is as misconceived as the thought that it must take longer to paint a slow-moving figure than it takes to paint a fast-moving one. There is no obvious reason why it should take one longer to imagine a figure rotating through 90° than to imagine it rotating through 45° - for one is at liberty to imagine the first figure rotating fast and the second more slowly. To assume that it takes longer to match a figure rotated by 90° than to match one rotated by 45° because the figure is being rotated at constant velocity adds a further incoherent hypothesis to the misconception. One might with equal cogency hypothesize that the gravitational force in mental space is also constant and that it is equivalent to 1G. The cognitive scientific research we have been discussing rests on the fundamental misconception that mental images are kinds of things that exist in a private mental space, which can be observed by the subject alone, and which he can move at will. It presupposes that

198

Human Faculties and Contemporary Neuroscience

when one imagines something moving quickly or slowly, then there is something thaimoves quickly or slowly in one's imagination, and further, that if something move-! quickly in one's imagination, then there is something that moves quickly. If one fails to grasp the structure of our concepts in this domain, then one may embark on investigations into the 'mental physics' of these putative 'mental objects', and suppose (with Finke) that mental images 'are governed by the same laws of motion' as physical objects. This is aii exemplary case of being caught in the trammels of language, enmeshed in the web of grammar.

7

Emotion

7.1 Affections The affections can be roughly subdivided into emotions, agitations and moods. These shade off into attitudes that are not emotions, such as liking and disliking, approval and ihaipproval, on the one hand, and into character traits, such as benevolence, irascibility and vindictiveness, on the other. The emotions are traditionally referred to as 'passions', inasmuch as one is, in an important sense, passive in their reception. They are not actions or, for the most part, even things one does, but things one feels, that one may be in the grip of or full of, that come over one and often overcome one. One can order a person to perform an action; but in so far as one cannot order someone to do something that is not a voluntary action, one cannot literally order someone to love or hate. One can decide and intend to act, but one cannot decide or intend to be angry or to feel jealous. Nor can one try to feel, succeed in feeling, or get better at feeling angry, love or pity (only become more irascible, loving or compassionate). On the other hand, we do say of a person that he ought to feel grateful or ought not to feel jealous, that his anger is warranted in the circumstances or that he has reason to feel resentful, and we blame people for their excessive anger or unreasonable resentment. Although we cannot feel emotions at will or to order, we can cultivate and refine our emotional responses. We can give way to our emotions or bring them under control, we can give expression to them or suppress them. This suggests that the idea that our emotions are not, in any sense, voluntary may be wrong. Indeed, we are sometimes, in some ways and to some degree, responsible, answerable, for our emotions. We shall investigate this matter below. emotions are a subclass of ions

Affections ave feelings. One can be said to feel love or Affections are feelings, but categori- , , . , „ , , ., . , . . . „ ,. . , ? ,. , hate (emotions), to leei excited or astonished (agitations), ally distinct from feelings that are , . , , ' , , , ... .. and to feel cheerful or depressed (moods). But the feelscnsattons, tactile perceptions or . . appetites * n § s ^ a t are affections are categorially distinct from the feelings that are sensations, which, unlike affections, have a bodily location and may inform one about the state of one's body. They are similarly distinct from the feelings that are modes of tactile perception, which, unlike affections, enable one to detect or apprehend features of one's environment. And they are distinct

200

Human Faculties and Contemporary

Neuroscieftce

from the feelings that are appetites, such as feeling hungry, thirsty or lustful, and from addictive feelings, such as craving for opiates or alcohol, that are a subcategory of appetites — namely, induced, non-natural, ones. They are similarly distinguishable from feelings that resemble appetites in certain respects, such as feelings of weariness, lassitude or fatigue These are not traditionally conceived of as appetites, perhaps because hunger, thirst and lust arise independendy of action, and lead to action (e.g. food-seeking behaviour), whereas weariness results from action and leads to inaction. „ , The distinction between affections in general and emotions r , Lht importance of distinguishing . . . , , , , , . „ „ f in particular, on the one hand, and appetites, on the other í.í r r rr 1 appetites jrom emotions - Rolls s ' is important to observe. Recent work on appetites has mor been mischaracterized as a result of failure to note the difference. E. T. Roils's book The Brain and Emotion purports to be an investigation into the neural substrate of the emotions. But it takes as its paradigmatic examples of emotions and as the object of all its experimental research, thirst, hunger and lust. Despite the interest of the results of these investigations, they are not about, and do not obviously have any bearing on, the emotions, for the simple reason that hunger, thirst and lust are appetites, not emotions. Appetites are blends of sensation and desire that are charAppetttes are blends of sensation . . r . . _, . , , , , . actenstic of animals. The sensations that are partly conr J ana desire . stitutive of appetites have a specific localization. The sensation characteristic of hunger is located in the midriff— one couid not have a feeling of hunger in one's throat, any more than one could have feelings of thirst in one's midriff. The feelings of hunger in one's belly must be distinguished from mere accompanying sensations such as light-headedness and dizziness that may occur after one has been hungry for a time. The sensation characteristic of thirst is the feeling of a dry throat that is, as it were, blended with a desire for drink. The sensations associated with appetites are forms of unease that dispose one to action to satisfy the appetite. The intensification of the sensations, particularly in the cases of hunger and thirst (as in the case of appetites of addiction), is progressively more and more unpleasant. 'Hunger', as Beaumont and Fletcher remarked, 'is sharper than the sword,' and a 'ravening fellow has a wolf in his belly'; but, as Eliza Cook observed, 'Hunger is bitter, but the worst of human pangs, the most accursed of Want's fell scorpions, is Thirst.' The desire that is blended with the sensation is characterized by its formal object. Hunger is a desire for food, thirst a desire for drink, and lust is a desire for sexual intercourse. What distinguishes the desires that are partly constitutive of appetites from other kinds of desire is not merely the fact that the former are blended with characteristic sensations, but also that the appetites lack any specific object. The child who announces that he is not hungry for the main course, but only for the pudding, is inadvertently making a grammatical joke. The adult who announces that he is very thirsty for a gin and tonic, but not for a cup of tea, is intentionally making one. The intensity of the desire is typically proportional to the intensity of the sensation. Satisfying an appetite leads to its temporary satiation and so to the disappearance of the sensation. Of course, the glutton may still want food, but no longer because he is hungry,

Emotion

201

just as the drunkard may want another drink, but not because he is thirsty. Appetites not constant, but recurrent. They are typically and naturally caused by bodily needs for, in £ n e c a s e of lust, by honnonally determined drives) consequent upon, for example, deprivation of food, drink or sexual intercourse, although needing food or drink or having a felt urge to have sexual intercourse is not the same as wanting it. Appetites are therefore unlike emotions. First, emotions Differences between appetites and . , ^•U , are not linked to localized sensations in the same way. ett Some emotions are associated with sensations {fear, rage), others are not (pride, remorse, envy). One does not have a feeling of pride in one's stomach or in one's chest, and although there are sensations characteristic of occurrent anger, such as throbbing temples and tension, one does not feel anger in one's temples or stomach muscles as one feels hunger in one's belly. Second, emotions have not only formal objects, in the sense that what one fears is what is thought to be frightening or harmful and what one feels remorseful about is a misdemeanour one has committed, they have specific objects, as when one fears tomorrow's examination or feels remorseful about lying to Daisy. Third, the intensity of emotions, as we shall see, is not proportional to the intensity of whatever sensations may occompany their occurrent manifestation. H o w much I fear heights may be manifest in the lengths I go to avoid them, and how proud I am of my children's achievements cannot be measured by reference to sensations. Fourth, emotions do not display the pattern of occurrence, satiation and recurrence characteristic of appetites, for the obvious reason that they do not have the same kind of physiological and hormonal basis as the appetites. Fifth, the emotions have a cognitive dimension absent from the appetites. The hungry animal wants food, the thirsty animal wants drink, the animal on heat wants sexual intercourse, but no particular knowledge or beliefs are essentially associated with these appetites. By contrast, the frightened animal is afraid of something it knows or thinks is dangerous; a mother is proud of her offspring inasmuch as she believes them to have such-and-such merits; a repentant sinner is remorseful, knowing himself to have done wrong. Finally, many human emotions are exhibited by characteristic facial expressions and manifested in typical tones of voice — as in the case of fear, anger, love and affection. Appetites are not. Paradigmatic emotions are things such as love, hate, hope, fear, anger, gratitude, resentment, indignation, envy, jealousy, pity, compassion and grief, as well as emotions of self-assessment such as pride, shame, humiliation, regret, remorse and guilt. Agitations are short-term affective disturbances, typically Agitations distinguished from ,, ,. , „,, . . . . caused by something unexpected. They include such temporary states as being and feeling excited, thrilled, shocked, convulsed, amazed, surprised, startled, horrified, revolted, disgusted or delighted. They are caused by what we perceive, learn or realize. Because they are disturbances, caused by unanticipated disruptions, they are not motives for action as emotions may be, but temporarily inhibit motivated action. One may behave in certain ways because one is excited, thrilled or shocked. But one does not act out of excitement, thrill or shock in the sense in which one acts out of love, compassion or gratitude. Agitations are modes of reaction: one cries out in horror or amazement, recoils with revulsion or disgust, is convulsed with ar e

202

Human Faculties and Contemporary Neuroscience

laughter or paralysed with shock. Occurrendy felt emotions, by contrast with lonaestanding emotional attitudes, often bear a kinship to agitations in the perturbations of f¡ example, the throbbing temples of rage, the trembling, sweating and shallow breathing i fear, the tears and cries of grief , , , , ,r Moods are such things as feeling cheerful, euphoric cop Moods distinguished from , . . ,5 , , ,. , , , tented, irritable, melancholic or depressed; they! are stativ emotions . or frames of mind one is in, as when one is in a state < melancholia, or in a jovial or relaxed frame of mind. They may be occurrent states i mind or longer-term dispositional states. One may feel depressed, melancholic, joyfL jovial, irritable or cheerful for an afternoon, or one may be suffering from a long-tendepression that lasts for months, as one may be in a cheerful mood for days on end. J1 a disposition, a mood is a proneness to feel, during one's waking hours, joyful, { depressed, or cheerful, and so forth. Moods are less closely tied to objects than emotion for one may feel cheerful or depressed without one's mood being directed at any specif object, whereas one cannot feel love without feeling love for someone or somethii or feel angry without feeling angry with anyone or about anything. Equally, mooi are closely linked not to specific patterns of intentional action, let alone to motive but to manners of behaviour. Cheerfulness, melancholia and depression, unlike lov envy or compassion, do not provide motives for action, but they are exhibited in tl manner in which one does whatever one does, in one's demeanour and tone of voic This is a corollary of the fact that moods colour one's thoughts and pervade one reflections. It is unwarranted, therefore, to characterize moods, as Moods are not emotional states „ , , , r , 1 1 Damasio does, as emotional states that are frequent or r that are frequent or prolonged , ^ continuous over long periods of time. One may fear war for a long period, but this does not imply that one is in any particular mood, although, to be sure, one's fear of an impending war may contribute to one's melancholic mood. Othello's jealousy was persistent and continuous, but, unlike his consequent depression, it was not a mood. One may be envious of A's success for years, but envy, whether prolonged or not, is no mood. And frequently, to fear things may be to be timorous by nature, but it is not to be in any mood. The boundaries between emotion, agitation and mood are not sharp. Emotional perturbations (as we shall refer to the typical somatic, expressive and behavioral manifestations of many occurrent emotions) have, as remarked, an affinity with agitations. Emotions may fade into moods, as when terror that abates leaves behind a mood of objectless anxiety. And conversely, a feeling of undirected anxiety may crystallize into a specific fear. The psychological category of the affections displays both conceptual complexity and diversity; the conceptual patterns to be discerned are irregular, and the variations from type to type are considerable. Consequendy, most generalizations concerning the concepts within the three subcategories need to be qualified with a 'for the most part' or a

'typically'. 1

A. Damasio, Tiie Feeling of What Happens (Heinemann, London, 1999), p. 341.

Emotion

203

7.2 T h e E m o t i o n s : A Preliminary Analytical Survey O n e m u s t distinguish b e t w e e n e m o t i o n a l character traits emotional character traits distin, •,• \. „ r i - \ • • J.. i, £"""' , (which are n o t reelings), e m o t i o n s as episodic p e r t u r b a mñshed from emotions as episodic , . , ,. : ., , &" , . , „., , tions, and e m o t i o n s as longer-standing attitudes ( b o t h perturbations and as attitudes ,- , -, , r r x w c r of w h i c h are said to be reelings). M a n y e m o t i o n terms have a use as names of character traits: w e speak of people as having a compassionate or

loving n a t u r e , as b e i n g o f a j e a l o u s or envious disposition, or as b e i n g irascible, t i m i d

or timorous by nature. Ascription of such character traits signifies a proneness to b e and to feel compassionate or loving, jealous or envious, angry or timid, given appropriate circumstances; a n d so too to act out of compassion or love, jealousy or envy, in anger or with timidity. T h e n o t i o n of felt emotion does n o t discriminate b e t w e e n Evisodic emotional perturbation dis- ,. , , , , __ ,y , , r . , . , an episodic e m o t i o n a l perturbation a n d a longer-standing tbwuished from emotional attitude , , _ . , _ . „ A . , . „ , , ,, e m o t i o n a l attitude. O n e m a y reel a w a v e ot pity t o r p o o r - neuroscience typically neglects the . \ i , • , so-and-so, feel funous w i t h s o m e o n e w h o has given offence, feel p r o u d to b e given the t r o p h y o n e is b e i n g awarded - here e m o t i o n clearly converges o n the subcategory of agitations. B u t equally, one may feel pity for s o m e o n e for as l o n g as h e is in a pickle, feel angry for years w i t h someone (but n o t furious or enraged - w h i c h are perturbational forms of anger), a n d feel proud for the rest of one's days to have w o n a trophy. 2 W h e n w e speak of a p e r s o n as 'giving way to his e m o t i o n s ' , 'controlling his e m o t i o n s ' , ' b e i n g o v e r c o m e w i t h e m o t i o n ' , it is usually e m o t i o n a l perturbations that w e have i n m i n d . W h e n w e characterize a p e r s o n as being ' e m o t i o n a l ' , w e d o n o t m e a n that h e feels love or hatred for m a n y p e o p l e , harbours n u m e r o u s fears a n d hopes, etc. R a t h e r , w h a t w e have in m i n d is that h e is p r o n e to emotional perturbation, is given to outbursts of feeling, expresses his anger, indignation, love or hate freely, a n d perhaps to excess, and tends to allow his e m o t i o n s to affect his judgement deleteriously. It is i m p o r t a n t , h o w e v e r , n o t to let this aspect of the e m o t i o n s occlude others, or to think that research o n e m o t i o n a l perturbation alone can p r o v i d e an adequate a c c o u n t of the e m o t i o n s in question. N e u r o s c i e n t i n e w o r k , m u c h influenced b y the misconceived J a m e s - L a n g e t h e o r y of t h e e m o t i o n s , has systematically, b u t unwittingly, screened o u t the attitudinal, as well as the motivational, cogitative a n d fantasy aspects of the emotions.

It is perhaps tempting to suppose that the term 'feeling' (as in 'feeling angry, afraid, proud') is confined to emotional perturbations, while 'being' (as in 'being angry, afraid, proud') eamiarks the emotional attitude. But that would be a mistake. For the most part, 'feeling angry' and 'being angry' are intersubstitutable. By and large, there is litde, if any, difference between feeling afraid of the impending war and being afraid of it, between feeling proud to have won the trophy and being proud to have won it. Respect and admiration, for example, involve no agitation, yet are nevertheless said to be fek, and there seems to be no difference between respecting and admiring someone and feeling respect and admiration for him. Where there is a nuanced difference in usage here, it is not to distinguish between emotional perturbation and emotional attitude.

204

Human Faculties and Contemporary Neuroscience

The standing emotional attitude is no less an aspect of our complex notion of an emotion than the correlative emotional perturbation. One's judgement may be clouded by one's emotional distress and agitation, as when one feels overwhelmed with grief, engulfed with rage, or in the grip of fear. But it may be equally distorted by one's long-standing resentments, envies or jealousies. Love of another person is an emotion which, on the one hand may be felt as an emotional perturbation — the wave of tenderness, the melting heart, the trembling hands, the blushes and so forth that may come over a young Romeo in the presence of his Juliet, and, on the other hand, as a standing attitude, exhibited, for example in the mature conjugal love of Pierre and Natasha in the Epilogue of War and Peace. The standing attitude is not a disposition to the corresponding episodes of emotional perturbation, but a lasting concern for the object of love, a standing motive for action beneficial to or protective of the person loved, a knowledge of and desire for shared experience, and a persistent colouring of thought, imagination and wish. The standing emotional attitude of love of, say, Pierre for Natasha is indeed linked to emotional perturbations, but not those of the melting heart, the joyful flushing, trembling hands and confused speech of youthful passion. Rather, it is linked to those perturbations that characterize concern for those one loves — anxiety about their welfare, joy in their achievement, longing to be reunited after absence, and so forth. (And one should not forget that one may love not only people (parents, spouse, children, friends), but also activities (conducting an orchestra, gardening, fighting battles), objects, places, views and landscapes (flowers, trees, the sea, one's home, one's city), works of art ( paintings, musical compositions), and ideals (honour, liberty) many of which are wholly unsuited to some of the occurrent perturbations of love.) Again, the hatred that Edmond Dantés bears his treacherous erstwhile friends is manifest far less in emotional outbursts than in the iron determination of his will for revenge, in the cast of his thought and fantasy, and in his motives and reasons for action over many years. The envy that moves Cousine Bette need not sweep over her every morning, but it informs her life in a multitude of ways. The love, hate or envy he in the manner in which the object of one's emotion matters to one and the reasons for which it is important for one - one cannot at the same time feel an emotion which is directed at a given object and also feel indifferent about that object. Hence, too, it lies in the motives that move one to action; for someone who feeis these emotions will normally act out of love, hate or envy. One's emotions are then evident in the reasons that weigh with one in one's deliberations, in the desires one harbours, and in the thoughts that cross one's mind in connection with the objects of one's feelings. One's emotions are inseparable from one's fantasy life and imagination, one's wishes and longings. So, for example, one dwells on one's hopes and fears, fantasizes about the fulfilment of one's wishes with respect to the objects of one's love or pride, and worries obsessively about the objects of one's anxieties. As noted above, one cannot measure a person's emotion , , , r i r i simply by the frequency or intensity of the emotional , * . r r r i perturbations he feels. A persons fear, for example, may be manifest above all in the lengths he will go to avoid situations which strike terror in his heart. The strength of its motivating force does not lend itself to quantification in the sense in which the rise in pulse, breathing, or perspiration rate does. Rather, its strength is evaluated by reference to the extent to which the emotion Emotion cannot be measured simply , , r , . . r by the frequency and intensity oj nprturhaf

Emotion

205

determines behaviour over time, and the kind of behaviour it determines. Moreover, although fear is a ubiquitous aspect of the animal, and hence too human, condition, it is a poor representative. For there are many emotions which typically involve little, if any, emotional perturbation or disturbance; for example, humility, respect, admiration, contempt and gratitude. Indeed, not at! instances of fear need involve fearful agitation, not because the fear is slight, but because of the character of the object of fear. What one is afraid of may preclude any particular, or at least any intense, fearful perturbation. Fear of imminent physical danger, harm or injury typically will. But fear of rain during the garden party one has planned for tomorrow, fear of a rise in the rate of inflation next month or of global warming over the next decades will not — even though the ensuing motivation may be powerful, and the effect on the agent's mood may be substantial. Similarly, the depth of a person's remorse may be exhibited not in a syndrome of sensations and perturbations that he feels, but rather in his strenuous endeavours to make amends for his past action and in his obsessive thoughts about his sin. The strength of a person's love for another is not to be evaluated only by the degree of his agitation in the presence of his beloved, but also by his concern for her welfare and by the sacrifices he is willing to make for her sake. The former without the latter may be more indicative of infatuation and desire than love. . ., Consequently, the idea of the duration of an emotion is r Ambiguity of duration oj , . , . . , . , , . r r r . , likewise equivocal. For it may refer to the duration of the perturbation of an emotional episode, or to the duration of a thought-infusing, fantasy-inducing, motivating emotional attitude which, in many different ways, may inform a person's life over a prolonged period of time. Any experimental investigation of the emotions must take Because of the conceptual complex- . , , . , _ , ;f. , / . , into account the complexity of the concept of an emotion ity and diversity of emotions, there ;. ' • , '. , , , and the conceptual diversity of the various emotions to r is no single conceptual prototype . ' which human beings are susceptible. There is no single paradigm of an emotion which can serve, as it were, as a conceptual prototype. Some emotions, when felt in appropriate circumstances and with respect to a certain range of objects, are closely associated with a pattern of sensations and characteristic facial expressions, as in the case of fear of physical harm, grief or anger. Others, such as hope, remorse or compassion, are not. Some emotions are exhibited in forms of agitation involving characteristic reactive behaviour, as in the case of terror (crying out, trembhng, turning white) or grief (weeping, wailing). Others, such as pride, respect or compassion are not. Some emotions are closely associated with relatively specific forms of action or inclinations to act, as in the case of fear of imminent danger (inclination to avoid or flee) or pity (inclination to help). Others, such as regret or hope, are not. So one should not take a single emotion, such as fear (let alone episodes of conditioned fear), restricted to a single object (e.g. pain), as a representative paradigm and generalize from that one kind of case. Conditioning rats to fear an electric shock (c£. LeDoux's important research) is a poor foundation for insight into human emotion and its neurophysiological conditions and accompaniments. ™. r Since so much experimental work is done on nonIhe differences between human and , . . . . . , ,.,,.„, „ • , . . human animals, it is important to be aware of the ditferr r anunal emotion are important jor neuwscience enees between human emotions, on the one hand, and the emotions of non-language-using animals, on the other.

206

Human Faculties and Contemporary Neuroscience

We share many emotions with other animals — for example, curiosity, fear and anger. I the scope of possible objects of human curiosity, fear or anger vasdy outstrips that of m animal curiosity, fear and anger. The cognitive and appraisive aspects of human emotit reach far wider than those of animal emotions. Of course, the monkey that screams in f of a snake knows snakes to be harmful. The lion that snarls in anger at the cub thai pestering it apprehends the cub as annoying. The pet cat that purrs with delight as its fc is being prepared knows that it is about to be given its meal. But the cognitive capacii of animals are stricdy limited by their lack of language. Many of the kinds of beliefs t enter into an account of human emotion are not beliefs that could possibly be ascribed an animal. Emotions, we have suggested, are ways in which we manifest what is import to us. But human beings characteristically reflect on what matters to them, whereas nc language-using animals merely manifest what they care about {their territory, possession the prey they have killed, their dominance in their group, etc.) in their behaviour. Hei the motivating power of an emotion in a non-language-using animal is both very restric and importandy different from the motivating force of emotions among human beir For human beings act for reasons, whereas animals at the most do so in a limited ; attenuated sense. Human emotions colour thoughts and fantasies, but non-language-us animals do not dwell in thought upon the objects of their emotions and do not fanta; about the fulfilment of their hopes and fears - they lack the conceptual equipment whicti makes such thoughts and fantasies possible. Finally, there are many emotions, such as feelings of guilt, awe, remorse and moral indignation, which it is not logically possible for non-language-using animals to have. For such emotions presuppose mastery of a language and possession of appropriate concepts. An awareness of the limitations on animal emotion and its objects should provide a brake on unwarranted generalization in which conclusions derived from experiments on animals are extended without more ado to human beings. So neuroscientific research that focuses upon conditioned fear in rats screens out most of what is distinctive of human fears, in particular, and emotions in general. Even if this is the correct point from which to begin one's endeavours to understand the neurological underpinings of human emotion, it is not at all obviously a point from which sweeping generalizations can be drawn — just as the calls of a wild animal are a poor basis for generalizing about the nature of human speech. Emotions generally have objects. If one is afraid, one is ,. , , , r . , r afraid ot someone or something, or that something has , . . . ... ,-, % happened or is going to happen; if one feels remorse, guilt or regret, it is for doing something; if one feels envy, it is envy of another for his good fortune. Evidently, there is more than one sense of'object of an emotion' at work here, and it is important, as we shall see, to distinguish them. The object of an emotion must be distinguished from its cause — what makes one jealous is not the same as what one is jealous of; your indignant tirade may make me feel ashamed, but what I am ashamed of is my own misbehaviour; a change in the fortunes of war may make one feel hopeful, but what one hopes for is final victory. Emotions are also linked in complex ways to knowledge Emotions ¡inked to knowledge and , , ,. r „ . , , r , a ... _ and belief. One cannot feel envious unless one knows or believes that another has enjoyed good fortune, feel Object and cause of emotion „ , ., , distinguished

Emotion

207

Valous unless one knows or believes that a person one loves is showing favours to another, feel fear of something unless one apprehends it as a threat to an interest one has. In this waV th e emotions essentially involve an appraisal or evaluation of their objects relative to the concerns of the agent. Knowledge or belief are equally involved in a human being's rationale for his emotion - that is,(in the specific reasons he has for feeling what he feels. Some emotions have characteristic somatic accompaniSotnatic p merits, sensations and physiological reactions that occur . . , . , , . _ , r v;„ml manifestations of emotions Vf on occasions in which the emotion — for example, tear, an ger or grief- is felt as a perturbation. Some emotional perturbations have characteristic behavioural manifestations. Some of these are expressive reactions of facial grimace, gesture a n d bodily demeanour. Others are voluntary expressive (as opposed to instrumental) actions; they are typically done not intentionally, for a purpose, or for the sake of a goal, and to the extent that they are, to that extent they are not purely expressive. Expressive manifestations of emotional perturbations include the manner of acting, even when the action is instrumental: when one is angry, one may slam the door rather than shut it quietly; when one is excited, one may raise one's voice; and one's hands may shake with fear during a viva voce. But standing emotional attitudes may also be exhibited in facial expression, mode of action, and demeanour — as is evident in the reciprocal behaviour of people who love each other, or the manner of behaviour of a person in the presence of another whom he admires and respects or fears and hates. Emotions commonly involve desire or aversion, and hence Emotions linked to volition, i-ij i- • . • ^ • 1-1. . . ,. ,£ are linked to volition and motivation. Emotions are linked motivation, rationality and jantasy to reasonableness and irrationality inasmuch as there can be reasons for an emotion, and the emotions commonly involve reasons for action for the the agent. Emotions are ¡inked to the imagination or fantasy, inasmuch as an emotion and its objects may inform one's thoughts, occupy one's daydreams, and preoccupy one during one's sleepless nights.

7.2.1 Neuroscientists' confusions W e shail explore a small part of this densely woven con, . . _ . , , , , , , -, ceptual web in a moment. But it should already be evident that the conception of emotion shared by eminent neuroscientists is flawed, partly as a consequence of an inadequate grasp of the concept of an emotion. So, for example, LeDoux summarizes the conclusions of his major book on the subject with the following observation: LeDoux s misconceptions

Emotional feelings result when we become consciously aware that an emotion system of the brain is active. Any organism that has consciousness has feelings. However, feelings will be different in a brain that can classify the world linguistically and categorize experiences in words than in a brain that cannot. The difference between fear, anxiety, terror, apprehension, and the like would not be possible without language. At the same time, none of these words would have any point if it were not for the existence of an underlying emotion system that generates brain states and bodily expressions to which these words apply. Emotions evolved not as conscious feelings, linguistically differentiated or not, but as brain states and bodily responses. The brain states and bodily responses

208

Human

Faculties and Contemporary

Neuroscience

are the fundamental facts of an emotion, and the conscious feelings are the frills that have added icing to the emotional cake. B u t this is confused, a n d does n o t provide insight i n t o either the nature of or the empir conditions for feeling any given e m o t i o n . (i) A n e m o t i o n m a y i n d e e d 'result' in response ti ,, , t T r cause. Unquestionably the appropriate functioning of ° b r a i n is a causal condition o f feeling e m o t i o n s . D a m _ " to tne ventromecua í prefrontal cortices, for example, to the somatosensory cortices in the right hemisphere severely compromises n o r m a l emotional responsiveness. B u t it w o u l d be misguided to assimilate the causal conditions for the possibility of feeling an e m o t i o n to the cause of a specific e m o t i o n o n a given occasion. So, for example, w h a t o n e is frightened by (e.g. Uns o u n d of a shot) is the cause of one's fear - n o t t h e c o n d i t i o n of one's brain that makes fear possible. F u r t h e r m o r e , the object of an e m o t i o n n e e d n o t b e the same as (although it m a y sometimes coincide with) the cause of an e m o t i o n . W h a t o n e is frightened by ~ namely, the sound of a shot — n e e d n o t b e the same as w h a t o n e is frightened of— namely, b e i n g killed — just as w h a t makes O t h e l l o jealous (Iago's fabrications) need n o t b e what lie is jealous of (Desdemona's supposed love for Cassio). W h e n o n e feels an e m o t i o n , one is not 'consciously aware that an e m o t i o n system of the brain is active'. For one feds e m o t i o n w i t h o u t k n o w i n g anything at all a b o u t the activities of the brain, any m o r e than o n e n e e d k n o w the cause (as o p p o s e d to t h e object) of one's e m o t i o n . T o feel love, jealousy or envy, a n d to be 'consciously a w a r e ' that o n e so feels, is n o t to b e aware of .in e m o t i o n system of the brain that is a causal c o n d i t i o n for being able to have such an e m o t i o n . R a t h e r , it is to feel love for so-and-so (jealous o r envious of so-and-so) and to realize that o n e does. , „ , (ii) W e have already n o t e d that the felt emotions of The differences between animal and , . c , . , a n o n - l a n e u a g e - u s m g animal are lar n a r r o w e r in ranee human emotion are not due to our , , i . , , .r • ,, • , a n d in obiect than the e m o t i o n s of a h u m a n beinii. I J brain s classifying things, but11to our ^ can possession of a language ' b u t m V d o S c a n n o t ' n o w h o P e for a S o o d d i r m e r o n Christmas day; I can, b u t m y d o g cannot, fear ihe ravages of war; and while m y d o g m a y b e apprehensive after having d o n e something it associates w i t h p u n i s h m e n t , it c a n n o t feel regret or r e m o r s e . T h e s e differences between h u m a n a n d animal emotions obtain n o t because o u r brain 'classifies the w o r l d linguistically a n d categorizes experiences in w o r d s ' (for this is n o t s o m e t h i n g brains can do), b u t becau-e w e (not o u r brains) have mastered a language, a n d that has enlarged the h o r i z o n of our t h o u g h t a n d feeling alike. (iii) It is m i s t a k e n t o suppose t h a t ' t h e difference Because animals have not mastered , , . , . ,, r , , , . b e t w e e n fear, anxiety, terror, apprehension and the like a language, they cannot have char. . ,. , ,. w o u l d n o t b e possible w i t h o u t language . T h e s e feelings actenstic human emotions r & & ¡r are n o t d e p e n d e n t o n mastery of a language. A cat may The causal conditions for feeling ... ., , r an emotion are distinguished jrom , i ( t, , if , . t r,, both the cause ana the object oj the emotion

3

J. LeDoux, The Emotional Brain, (Phoenix, London, 1998), p. 302.

Emotion

209

feel anxious when it perceives a large dog, even though the cat is safely up a tree; a dog „-aV exhibit fear when it is threatened with punishment; and a rabbit pursued by a fox m a y feel terror. But it is perfectly true that many emotions that human beings feel are not possible for non-human animals. Only a language-user can fear bankruptcy, feel anxious about inflation, feel terror of ghosts, and feel apprehensive about the state of the nation. These are emotions of a type which non-human animals can feel, but with objects which He beyond their cognitive horizon. Similarly, only a language-user can feel remorse for his Hes or treachery, feel awe at the sublime or contempt for the despicable. These are emotions of a type which non-human animals cannot feel. (iv) Emotions cannot be said to have evolved as 'brain , , ... , n , , . , . nFmotiotis are neither brain states '" . . states and bodilyJ responses . Rather, brains evolved in r nor somatic reactions . . such a way as to make it possible tor animals to respond affectively to objects of their concern. Emotions evolved as animals' responses to features of the environment apprehended as affecting in one way or another the good of the animal. Neither brain states (which are essential for feeling an emotion) nor somatic responses (which may characterize an emotional perturbation) are emotions. They lack the intentionality, or 'directedness towards an object', which is constitutive of most emotions. One cannot individuate an emotion by reference to either brain states or somatic reactions independently of the circumstances of their occurrence and the knowledge or beliefs, as well as the desires or wishes, of the creature. (v) Emotion words do not apply to brain states at all, Emotions ana somatic responses , , , , . , , - , • , but to creatures, who feel emotions and exhibit them in their behaviour. Emotion words are not names of bodily responses that characterize emotional perturbation. Brain states are 'the fundamental facts of an emotion' only in the sense that they are essential for the animal to feel the emotion it feels, as indeed they are also essential for the animal to breathe or move, to perceive and to respond to what is perceived. Bodily responses are not 'the fundamental facts of an emotion' in any illuminating sense. My acrophobia may be manifest in my avoidance of heights, not in my sensations when I climb heights - since I carefully avoid climbing them precisely because of my fear (once may have been quite enough). Anger at the Ruritanian National Party need not be manifest in going red in the face and shouting (such behaviour being unseemly), but in terminating one's subscription to the party. And love of honour in a knight errant was manifest not in perturbations of love, but in what he counted as a reason for action, in the lengths he was willing to go to preserve his honour, in his admiration for honourable men and their deeds, and in the perturbations of indignation at what he viewed as dishonourable in others. The bodily responses of fear of physical harm - for example, increased pulse rate, perspiration and trembling - can all be exhibited without any fear whatsoever, as when one trembles with excitement on entering a hot room expecting a delightful surprise. What make those responses fear responses are the circumstances in which they are exhibited, and the beliefs, desires and thoughts of the agent. The 'fundamental facts of an emotion', in the case of human beings, are the agent's awareness of, or beliefs about, an appropriate object of emotion in the circumstances, the character of his concern for the object of his emotion (why it matters to him), and the consequent reasons for action he may have, the motivation afforded the agent by the relevant appraisal or

210

Human Faculties and Contemporary

Neurosdence

evaluation, the behaviour or behavioural disposition thus connected with the object of tr emotion, and the associated thoughts, fancies and wishes. That a creature feels fear when apprehends danger, or anger when it perceives encroachment upon its territory; that human being feels hope that a certain desideratum will eventuate, feels filial, conjugal ( parental \ovt, feels pride at the completion of a difficult and worthy task, remorse for a si or embarrassment at a gaff, are not the 'icing [on] the emotional cake' but the flour fro which the cake is made. It is precisely in the context of the recognition of an appropria object of an emotion, of concern for it, and of a form of behaviour or inclination i behave appropriate to that object (given the agent's goals and beliefs) that somatic accon paniments, voluntary actions and involuntary reactions of an agent can be characterized manifestations of that emotion. (vi) It is a mistake to distinguish between having ; LeDoux is mistaken to think that . , r .. . , ° , ., emotion and feeling an emotion, and to suppose th a creature can have an emotion with. z. , r ,. •. animals that are conscious can both have and reel emi out feeling it " tions, whereas animals lacking consciousness may hai an emotion — that is, appropriate brain states and bodily responses — even though they c not feel emotions. Emotions, we have noted, are not brain states or bodily responst They are neither objects of perception nor sensations that are felt. T o feel frightened ju is to be frightened, just as feeling angry is not distinct from being angry. One cannot I afraid or angry on an occasion and not feel afraid or angry. It is a further confusion suppose that there are animals that can have emotions but that are not conscious ( opposed to self-conscious) creatures fa distinction that we shall examine in chapter 12) Comparable confusion is evident in the writings of anoth Damasio on the emotions ... .. , . . . . „ . , , distinguished neuroscientist. Antonio Damasio s work < patients suffering from emotionally incapacitating brain damage is righdy renowned, and his insistence on a link between the capacity for rational decision making and consequeni rational action in pursuit of goals, on the one hand, and the capacity for feeling emotions. on the other, is bold and thought-provoking. However, his speculations on the emotion-, are, in our view, vitiated by conceptual confusion. Damasio's conception of the emotions is much influenced The influence of fames on , -, w -n -r i i i 1 i .. • ^- 11 . by William James, who held that emotions are essentially n the feelings of somatic disturbances consequent on the perception of an 'exciting fact'. An emotion, according to James, is not the somatic change itself, but the agent's perception or apprehension of it. The changes are indefinitely numerous and subde, and the entire body is a 'sounding board' for such excitations. Every one of these bodily changes is allegedly felt as soon as it occurs. One cannot abstract from an emotion 'ail the feelings of its bodily symptoms' and find anything left behind other than a cold and neutral state of intellectual perception. 4 Damasio contends that James. 'well ahead of both his time and ours, . . . seized upon the mechanism essential to the 4

William James, The Principles of Psychology (Mok, New York, 1890), vol. 2, pp. 449-51. For a brief critical discussion of James's confusions, see A. J. P. Kenny, Action, Emotion and the Will (Routledtio and Kegan Paul, London, 1963), pp. 39-41.

Emotion

211

derstanding of emotion and feeling'.3 Damasio himself sees 'the essence of emotion as collection of changes in body state that are induced in myriad organs by nerve cell mimáis, under the control of a dedicated brain system, which is responding to the ntent of thoughts relative to a particular entity or event1.6 The somatic changes are held be caused by thoughts. Damasio's conception of thoughts is firmly rooted in the piffhteenth-century empiricist tradition. Thoughts, he claims, consist of mental images (which may be visual or auditory, etc., and may be of items in the world or of words or svrnbols that signify such items). 7 The images constituting thoughts are comparable to the images of which perception allegedly consists, differing from them in being fainter or less lively- in this respect Damasio self-consciously but misguidedly follows in the footsteps of Oavid Hume. 8 Damasio apparently holds the view that if thought were not exhibited to us in the form of images of things and images of words signifying things, then we would not be able to say what we think. 9 . , . Damasio, unlike James (but like LeDoux), distinguishes Damasw P J _ that is, 'a collection of changes in body state m emotjon connected to particular mental images that have activated a specific brain system' — from the feeling of an emotion. 'The essence offeeling an emotion is the experience of such changes in juxtaposition to the mental images that initiated the cycle. In other words, a feeling depends on the juxtaposition of an image of the body proper to an image of something else, such as the visual image of a face or the auditory image of a melody.' 10 So, an emotion is a bodily response to a mental image, and the feeling of an emotion is a cognitive response to that bodily condition, a cognitive response 'in connection to the object that excited it, the realization of the nexus between object and emotional body state'.11 Feelings of emotion,

5

Antonio R. Damasio, Descartes' Error: Emotion, Reason and the Human Brain (Papermac, London, 1996), p. 129. He remarks that James's insights on the human mind have been rivalled only by Shakespeare's and Freud's. This praise, in our view, would be more properly directed towards William's brother Henry. 6 Ibid., p. 139. 7 Ibid, pp. 107f. s Ibid., p. 108. Hume's account is in A Treatise of Human Nature, I. i. 1. 9 He writes: 'Most of the words we use in our inner speech, before speaking or writing a sentence, exist as auditory or visual images in our consciousness. If they did not become images, however fleetnigly, they would not be anything we could know' (Descartes' Error, p. 106). He fails to see that one LOUU not find out what one thinks by mere attention to whatever mental images cross one's mind when one thinks. A mental image may, as it were, illustrate a thought, as a picture may illustrate a text. But it is the thought that makes the mental image the image of what it depicts, just as it is the text of the book that makes the illustration an illustration of the story. Without the text the picture of Lancelot in full armour could depict any knight whatsoever, illustrate how to sit in the saddle, or how not to sit in the saddle, illustrate armour of a given period, or depict the kind of horse used by knights in annour, and so on. And what applies to the relation of picture to text applies similarly to the relation of mental image and thought. 10 11

Ibid., p. 145. Ibid., p. 132.

212

Human

Faculties and Contemporary

Neurosáence

Damasio avers, 'are just as cognitive as any other perceptual image, a n d just as dependent < cerebral-cortex processing as any other image'. H o w e v e r , feelings are about something different. But what makes them different is that they are first and foremost about the body, they offer us the cognition of our visceral and musculoskeletal state as it becomes affected by preorganized mechanisms and by the cognitive structures w e have developed under their influence. Feelings let us mind the body . . . Feelings offer us a glimpse, of what goes on in our flesh, as a momentary image of that flesh is juxtaposed to the images of other objects and situations; in so doing, feelings modify our comprehensive notion of those other objects and situations. By dint of juxtaposition, body images give other images a quality of goodness or badness, of pleasure or pain. 12 Accordingly, D a m a s i o proposes w h a t h e calls 'the soma , , , . , _, . . . . . m a r k e r hypothesis . T h e hypothesis is that somatic j Jr . t sponses to 'images' (i.e. perceptions and thoughts) ser to increase the accuracy and efficiency of decision processes, screening o u t a range alternatives a n d allowing the agent to choose from a m o n g fewer. 1 3 ' W h e n a negati somatic m a r k e r is j u x t a p o s e d to a particular future o u t c o m e the c o m b i n a t i o n functions an alarm bell. W h e n a positive somatic m a r k e r is j u x t a p o s e d instead, it b e c o m e s a beac of incentive.' 1 4 So somatic markers, constituted b y the somatic response to situations cc fronting us, assist deliberation b y highlighting s o m e options and eliminating t h e m . Thi somatic responses w h i c h w e allegedly use for decision m a k i n g 'probably w e r e created o u r brains during the process of education and socialization, b y c o n n e c t i n g specific clas of stimuli w i t h specific classes of somatic state'. 1 5 Culturally inculcated 'gut reactio p r o v i d e t h e basis for rational decision making. 1 6 T h i s leads Damasio to conjecture that t decision-making and executive deficiencies in patients suffering from lesions in the prefror cortices is to b e explained in terms of a lack of somatic markers to guide t h e m . This conception, w e suggest, involves conceptual confusion. Damasio s somatic marker , , . hypothesis

12

Ibid., p. 159. Ibid., p. 173. 14 Ibid., p. 174. 15 Ibid., p. 177. 16 Damasio's theory is to a large degree a stimulus—response theory, operating at the neural, rath than at the behavioural, level. He writes: 13

Somatic markers are thus acquired by experience, under the control of an internal preference system and under the influence of an external set of circumstances which include not only entities and events with which the organism must interact, but also social conventions and ethical rules. The neural basis for the internal preference system consists of mosdy innate regulatory dispositions, posed to ensure survival of the organism. Achieving survival coincides with the ultimate reduction of unpleasant body states and the attaining of homeostatic ones, i.e. functionally balanced biological states. The internal preference system is inherently biased to avoid pain, seek potential pleasure, and is probably pretuned for achieving these goals in social situations.' (Ibid., p. 179)

Emotion

An emotion is not an ensemble of somatic changes , , , , caused by a thought about (i.e. mental image of) an ; b . \ • , 'u object or event, hirst, even if a given emotional perturbaJ & . ^ & tion does involve a range of somatic changes, what makes • sensations sensations of fear as opposed to anger, and what makes the blushes blushes shame rather than of embarrassment or of love, is not the 'thought' or mental image, if any, that causes them, but the circumstances and the object of the emotion. One might indeed argue that what connects the sensations with the circumstances and object of the emotion is that, had the object of the emotion not been apprehended in the manner in which it was, the sensations would not have occurred — and perhaps this is part of what Damasio had in mind. Nevertheless, it is not the somatic changes or their apprehension that constitutes the emotion. Secondly, if emotions were essentially ensembles of somatic changes caused by thoughts (mental images) - that is, if that is what the term 'emotion' means — then learning the meaning of emotion words, and hence learning how to use them, would be a matter of learning the names of complexes of bodily changes with specific causes {akin to learning the meaning of an expression like 'giddiness' or 'seasickness'). But we do not learn the use of emotion words by learning sensation names or names of overall bodily condition, but rather by learning what are appropriate objects of the relevant emotions - for example, of fear (what is dangerous or threatening), of anger (what is annoying, offensive or in some way wrong), of pride (worthy achievement or possessions), of guilt (one's own moral misdemeanours), and so forth, and learning how to use these terms ('afraid', 'angry', etc.) in the expression of one's feehngs towards the appropriate objects and in the description of the feehngs (but not the sensations) of others. Thirdly, if emotions were ensembles of somatic changes caused by mental images, then one could not have good reasons for feeling a certain emotion, and would not be answerable for one's emotions in the manner in which we are. For although there may be a reason (i.e. an explanation) as to why one has a headache, or why one's breathing rate or heartbeat rises, one cannot have a reason (i.e. a ground or warrant) for such things. Given appropriate circumstances, we can say that someone ought to, and has good reason to, feel proud or ashamed, but we cannot say (save in what is merely a predictive sense) that his pulse rate ought to rise, or that his psychogalvanic reflex reactions ought to change. Fourthly, one can feel an emotion E without any E-type perturbation. One can love a person, an object (place, artefact or work of art) or a value without undergoing any somatic changes of love when one thinks about the parents, wife or children one loves, or about Venice, Chartres Cathedral or Beethoven's late quartets, let alone about liberty, justice or honour. There need be no somatic changes accompanying the thought that the rate of inflation is likely to rise - but one may well fear that it will. One's pulses need not race m order for one to hope that tomorrow's picnic will be a success. If A did one a majoi favour in one's youth, one may remain grateful for the rest of one's days - but one need not break out in a sweat whenever one thinks of A and the favour done. One may be proud for the whole of one's subsequent life to have won an Oxford Blue, but there are no somatic changes characteristic of pride in something — or of many other types of emotion and many other emotions with certain kinds of object. tn.iüo's confusions: An emotion ffWi J ' , ,, t a somatic change caused by a lP " ^ ,. ,. uoht. Four objections

(i)

213

214

Human Faculties and Contemporary Neuroscience

Of course, to insist on these points is not to deny that there is an essential link betweei certain emotions and emotions directed to specific objects, on the one hand, and types o emotional agitation involving somatic changes, on the other. At the very least, the emo tional agitation may be characteristic of that emotion, or of that emotion with that type 0 object, given appropriate circumstances. It is merely to insist that the emotion is not th< somatic changes that might be caused by the thought of the object of such an emotion. (ii) It is, as we have already argued (§4.2.3), a mistaki Damasio s confusions: any somatic , . . ,. , , , :, to suppose that perceiving an object or perceiving tha changes need not be caused by meno « 1 t , . things are thus-and-so involves having images of any Lili I tit

ftíjcj

*

thing. It is equally misguided to suppose that in orde to think something or think of something, it is either necessary or sufficient to have ai image of anything, let alone an image of what one thinks or thinks of or of words tha would, if uttered, express what one thinks or refer to what one thinks of. This should bi evident from the schematic analysis of thinking in the previous chapter. Moreover, as wi shall argue in §12.5, it is mistaken to suppose that one thinks in images, or that in orde to speak with thought, one must first say to oneself in one's imagination what one is goinj to say out loud. One can talk to oneself in the imagination (which involves auditor images) without thinking (as when one counts sheep in the imagination or recites mantra in order to prevent oneself from thinking), and one can think without talking t< oneself in the imagination (as when one speaks thoughtfully to another, engages in ai activity with thought and concentration, etc.). Since neither thinking nor perceiving need involve images, the somatic changes tha may be part of a given emotional agitation and which may (but need not) be caused by thought (in the proper sense of the term) or by perceiving something need not be causei by mental images. (iii) While there is a difference between feeling an Damasio is mistaken in his distinc, . , . , ... , r ,. , . . , „ ,. emotion (e.g. feeling eaious) and realizing what emotion Hon between having and jeehng an ° , . ,. one feels (e.g. that it is jealousy), there is, by and large, as previously noted, no significant difference between having an emotion and feeling an emotion (being jealous and feeling jealous), any more than there is a difference between having a pain and feeling a pain. Where there is a subtle difference, it is not between the occurrence of somatic changes and the apprehension of such changes in association with mental imagery. Damasio's stipulated distinction between emotion and feeling an emotion has nothing to recommend it, since an emotion is not an ensemble of somatic changes, and feeling an emotion is not the experience of such changes in juxtaposition to mental images that caused them. (iv) It is mistaken to suppose that feeling an emotion An emotion is not a cognitive reis a cognitive response to a bodily condition caused b \ sponse to a bodily condition caused mental images. If, when frightened by a noise at night. by mental images I feel frightened that there is a thief in the house, and m\ pulses race, my (felt) fear is not a response to my racing pulse. What I was frightened by

Emotion

215

was the noise (not an image of a noise), w h a t I was frightened of was a burglar's having broken in (which m a y or m a y n o t have b e e n the case). I m a y or m a y n o t notice m y racing tmlses - but w h e t h e r I d o or d o not, m y fear of a burglary is n o t a response to t h e m .

emotions are not about the somatic £"" , , chames that accompany them " A

(v) Feelings of e m o t i o n are not about the body at all. "What , < , , . , ,. c , they are about is the obiect of the e m o t i o n , in o n e or J ' a n o t h e r sense or the t e r m (see below). W h a t a person is

proud about (or o f ) m a y b e his achievements, his lineage, his children, his possessions, etc. - but n o t any somatic changes that m a y o c c u r w h e n h e thinks o f t h e m . W h a t a p e r s o n feels guilty a b o u t are his sins or w r o n g d o i n g s , n o t any bodily perturbations that m a y or may not occur w h e n h e thinks a b o u t t h e m . W h a t a person feels angry about is perhaps the annoying b e h a v i o u r of another, b u t n o t (normally) his somatic responses to it. (vi) An emotional response n e e d n o t b e cognitively ,. . , , , . , - , r linked to the cause of the e m o t i o n or the cause of the , . somatic changes that m a y a c c o m p a n y an emotional p e r . ° . turbation. W e are often ignorant o f the causes of o u r emotional feelings. I m a y n o t k n o w w h a t caused m e to feel love for Maisy, to hate injustice, to feel fear of death — b u t w h a t I must k n o w , w h a t is 'cognitively linked' to these feelings, is w h a t their objects are. If the c o n n e c t i o n of a feeling of e m o t i o n to w h a t it is 'about' — that is, its object — w e r e a causal o n e , then unless such causal k n o w l e d g e were n o n - i n d u c t i v e , k n o w i n g w h a t o n e is frightened of, angry w i t h or about, p r o u d or ashamed of, w o u l d b e a hypothesis. B u t I d o n ' t discover the object of m y feelings by tracing t h e causes of t h e perturbations (if any) that I feel. (vii) O n e ' s 'feelings of e m o t i o n ' , one's love or hate, r , Emotions ate not ways of discovering r , ., , . _ ,. ,. r t , t +• r , tear or h o p e , pride or shame, are n o t ways of finding r r somatic jacts, but somatic jacts may ' ° o u t facts a b u t inform one about one's emotions ° ' o u r v i s c e r a l a n d musculoskeletal state'. Indeed, one's emotions d o n o t inform o n e of either the state of one's b o d y or the state of the w o r l d a r o u n d o n e . B u t one's emotional perturbations may inform o n e of one's e m o t i o n a l attitudes. A p a n g of jealousy m a y indicate that I a m in the process of falling in love w i t h Maisy; a blush of embarrassment m a y b r i n g h o m e to m e that I am ashamed of h a v i n g lied; m y tears of grief may m a k e m e realize h o w m u c h I loved Daisy. Far from one's e m o t i o n s informing one a b o u t the state of one's b o d y , the state of one's b o d y informs o n e a b o u t one's e m o t i o n s . Feeling grief does n o t inform m e of the state of m y lachrymal glands, b u t m y h o t tears m a y s h o w m e just h o w m u c h I grieve for so-and-so. Feeling fear o n a given occasion does n o t inform m e of the condition of m y heart, but m y t h u m p i n g heart m a y s h o w m e h o w frightened I am. Feeling embarrassed does n o t inform m e of the state of m y facial arteries, b u t m y blushes m a y indicate to m e that 1 a m m o r e embarrassed than I w o u l d have imagined. Emotional response is independent , , , , , , of knowledge of the cause of the J . ' emotton

~, , , lhe somatic marker hypothesis is . , misconceived

(viii) Damasio's somatic m a r k e r hypothesis is m i s c o n . , ,_, . . i n ceived. I h e e m o t i o n s are not somatic images that tell . ° o n e w h a t is g o o d and bad. Bodily reactions are n o t ersatz

216

Human Faculties and Contemporary

Neurostience

guides to what to do, and do not inform us about good and evil. If one is indignant at perceived injustice, what tells one that the object of one's indignation is an evil is not th one feels flushed in association with the thought of the act in question. O n the contrar one is indignant at A's action because it is unjust, not because one flushes in anger whc one hears of it. And one knows it to be unjust because it rides roughshod over someone rights, not because one flushes in anger. Indeed, the flush is only a flush of anger in so f as one is thus indignant. And one will feel indignant only to the extent that one c.ir about the protection of the rights of human beings (or of this human being). , , One might conjecture that although Damasio may 1 It is the capacity to care that con, . . . . r r . , ., . ,. perfectly correct m associating the capacity for ratiouali nects the emotions with rationality , _ m in pursuit of goals P m c t l c a l reasoning and m pursuit of goals witli if ability to feel emotions, the linkage lies in a conmic feature underlying both. Since the emotions do not let us 'mind the body', and MIII feeling the somatic reactions to circumstances is not a litmus test for good and evil, or i'. the beneficial and the harmful, it is implausible to suppose that what is wrong wi patients who have suffered damage to the ventromedial sector of the prefrontal cortex that their somatic responses are awry or uninformative for them (which would be. a-, were, a Pavlovian deficiency). But what might be investigated is whether the bra damage in the kinds of patient that Damasio studied affects the capacity to care or persist in caring about goals and objectives. For such a deficiency would affect both tl patients' emotions and their ability to pursue goals over time. One feels no emotio about things concerning which one is indifferent, and one does not pursue goals effiuenu, unless one cares, for one reason or another, about achieving them. So, clarification of the concept of emotion is relevant to neuroscientific investígaium (and so is clarification of the related concepts of mood and agitation). Unclarity .ibout these concepts is likely to generate incoherence in experimental investigation and confusion in the interpretation of experimental results. We shall try to shed some light on the matter by examining several issues already raised: the multivalent notion of the obicu of an emotion, and the distinction between object and cause; the connection of emotion to knowledge, belief, appraisal and concern or care; the somatic accompaniments and behavioural expression of an emotion; and the connection of emotion with motivated action.

7.2.2 Analysis of the emotions The concept of the object of an emotion needs clarifkjtion. For one must distinguish between different seines of'object' in this context. If one asks what the objjet of someone's love, fear or anger is, one might specify the person or thing one loves, fens or is angry with. The teacher, one would then suggest, is the object of one's fear, n«>i lib wrath; the man who won the lottery is the object of one's envy, not his winning the lottery; the benefactor, not the benefit, is the object of one's gratitude; and Jack. w\:h whom Jill is flirting, is the object of one's jealousy, not her flirting with him. This is one sense of the term 'object of an emotion' (which may be singled out as the object acci.-aln'i of an emotion verb). The object of our emotion, in this sense, is the referent of jit Different senses of 'object' of an emotion

Emotion

ropriate

217

referring expression. If o n e feels an e m o t i o n E w i t h respect to a certain object,

fhus understood, t h e n the object of the e m o t i o n m a y or m a y n o t exist, b u t the agent must, at any rate, believe that it does. In this sense o f ' o b j e c t ' , n o t every e m o t i o n n e e d have an object, and some e m o t i o n s c a n n o t have o n e (e.g. h o p e ) . Equally, o n e m a y fear disaster, h o p e for victory, feel pride, shame, regret or remorse for having d o n e s o m e t h i n g , or feel resentful or indignant that A has acted unfairly, feel envious of B's success. Believing that o n e has d o n e s o m e t h i n g u n t o w a r d , o n e m a y feel guilt, regret or remorse, even if one's belief is false, just as O t h e l l o m a y b e jealous of p e s d e m o n a ' s love for Cassio, e v e n t h o u g h she does n o t love Cassio. I n this sense o f 'obiect of e m o t i o n ' , the object is singled o u t by a nominalization accusative — that is, an abstract n o u n (e.g. 'disaster', 'victory', ' D e s d e m o n a ' s love') or a n o u n - c l a u s e (e.g. of the form 'that such-and-such is, was or will b e the case'), a n d it does n o t follow from the fact that one hopes for victory or fears that o n e will b e defeated that o n e will b e victorious o r that one will be defeated. 1 7 , , Wiiat the formal character oj the , . object of an emotion is

M a n y e m o t i o n s are partly defined b y reference to the , . , • ,• . n f t, t £ - , O n e c a n n o t feel remorse Jformal character of their objects. J . J tor h a v i n g d o n e s o m e t h i n g o n e k n o w s n o t to be w r o n g , for remorse is essentially directed at one's o w n past offences. O n e c a n n o t h o p e for something that o n e k n o w s has h a p p e n e d or c a n n o t happen, for h o p e is essentially directed towards a desired eventuality the o c c u r r e n c e of w h i c h is in d o u b t . O n e c a n n o t feel indignation or resentment, b u t only s h a m e , regret or remorse, at w h a t o n e has d o n e o r thinks one has d o n e oneself, for indignation a n d resentment are essentially directed at the actual or supposed misdeeds of others. T o fear s o m e t h i n g is to a p p r e h e n d the object of one's fear as dangerous or as a threat, and accordingly t o feel apprehensive. O n e c a n n o t envy another unless o n e sees that person as possessing a g o o d that o n e wishes to have oneself. So, if a person feels a certain e m o t i o n w i t h respect to a certain object, t h e n , for many emotions, h e m u s t h o l d the object of his e m o t i o n to satisfy t h e relevant formal requirements if it is i n d e e d t h e case that h e feels the e m o t i o n in question. C o n s i d e r a t i o n of the c o n c e p t of the formal object of an Tiie connection between emotion . , • , , 1 • 1 , , . e m o t i o n makes it clear that e m o t i o n s c o m m o n l y involve and evaluation ' an appraisal or evaluation. T h e manifestation of an e m o t i o n exhibits an appraisal of people, things a n d events relative to one's concerns (and one's c o n cerns may stretch far b e y o n d one's personal welfare a n d iHfare). Pride, shame, embarrassment, feelings of guilt and humiliation are obviously e m o t i o n s of self-appraisal. B u t an e l e m e n t of appraisal is generally involved in e m o t i o n s : fear involves the appraisal of a situation as threatening or dangerous; h o p e involves the evaluation of a possible situation as desirable; This still leaves many further grammatical constructions. One may love playing tennis, hate war, fear dying, and it is not obvious that these can be recast in the form of a that-clause. T o love playing tennis is not to love that one plays tennis (one should be wary of mangling grammar), to hate war is not the same as hating that anything, and to fear dying is not the same as fearing that one will die. These difficulties need to be resolved, but they merely confirm our point: viz. that the term 'object of an emotion' masks the logical diversity of the kinds of things that can be specified in answers to the question 'What (or whom) do you E?' or 'What are you E of (or at, etc.)?'.

218

Human Faculties and Contemporary Neuroidence

anger involves the appraisal of an act, event or agent as (in some way or other) w r o , envy an evaluation of another person's circumstances as desirable for oneself, and so for „ , , . . The object of one's emotion {in one sense or the orii U1 Only what one cares about (posit,. , „, r , , ,- / i i ,is something oí consequence tor one. That which r, wely or negatively) can be an object ° ^ "-11 0 l o v e s o r h a t e s fears o r h o e s ior feels of one's emotion ' P P r o u d o r asllarrj of, is something or someone that, in some way, matt to one (positively or negatively). One is anything but indifferent to the people and thii one loves or hates. The objects of one's hopes or fears are things which signify, thii which one knows or believes affect one's interests or concerns. What one is proud ashamed of is (very roughly) something the possession or performance of which c believes enhances or diminishes one's worth - and that is something we naturally c about. One cannot feel angry, resentful or indignant about something towards which c is altogether indifferent, any more than one can feel remorse for doing something t does not matter to one or guilt for doing something which does not rate. One would i be moved to act out of pity or compassion if one did not care for the object of or feelings, and one would not be impelled to act out of pride or shame if one did not c for one's self-esteem or for the esteem of others. When things cease to matter, one cej to feel anything about them. One's emotions are the various forms which one's cone may take. In manifesting our emotions, we show what we care about. The objects of < emotions and the intensity of our feelings about them reveal what sort of person we ar hence the non-contingent connection of the emotions with character traits. „ , , , . , Emotions are linked in complex ways to what the ag The link between emotion and . , , . „ . , . . , , , , ¡ vc knows or believes, bor m so far as an emotion must h, knowledge or belie) a proper object in order to qualify as the emotion it the agent must take the object of his emotion to satisfy the formal characteristics which det mine the object as appropriate. If he fears A or A's action, he must know or believe tn,u A or A's action is a threat. If an agent feels pity or compassion for another, he must know or believe that person to have suffered a misfortune. If he feels regret, remorse or guilt, he must know or believe that he has done something unfortunate, wrong or untoward; and if he feels envious of another, he must know or believe the other to be better off in some way than himself. In addition, the agent must, for many emotions, have an array of furthei beliefs regarding the object of his emotion (in either sense of the term), beliefs which he holds to warrant the ascription to the object of the appropriate formal characteristics. I-or the formal characteristics are shared by all objects of a given emotion; whatever A regret-» is something he wishes he had not done or had not had to do, but the reasons which maki? K'ing regrettable will typically be very different from the reasons that make X'ing regrettable. What a person fears is something he believes to threaten his interests, but his reasons for fearing tomorrow's exams (viz. his knowledge that he is ill-prepared and may fail) provide quite different grounds for seeing what he fears as threatening his interests from his reasons for fearing heights (viz. his belief that he may fall and injure himself). To the extent that his reasons are cogent, they may also provide a justification for his emotion.

1S But they may not: one might argue that envy, jealousy and hatred, for example, are rendeied intelligible by the relevant reasons, but that they are not justified.

Emotion

219

This cognitive component in emotions has important ramifications. First, a person cannot rlinarily feel a certain emotion and fail to know what the object of that emotion is. He not feel grateful and yet know neither to whom he is grateful nor what he is grateful - he cannot feel ashamed and yet not know what he is ashamed of doing; he cannot feel vet know neither for whom nor for what. In limiting cases, one may feel objectless (i.e. Angst), or irrational guilt which lacks an object, or longing without there being vthing determinate for which one longs; but these are necessarily exceptions to the rule. Secondly, even though emotions belong to the category ,„ rrmnection of the emotions . , . , . , . r

r / t? ' . The subjective somatic accompaniments consist in the sensations that commonly characterize a given emotional perturbation - for example, the sensation of beating pulses and thumping heart, feelings of tension, the 'butterflies in the stomach', the felt dry throat, that characterize fear, hope and excitement. The objective somatic accompaniments are the physiological changes that typically characterize the emotional perturbation, the neural excitation in the brain, visceral activities and glandular secretions, the psychogalvanic reflex, and so forth. The behavioural expressions of an emotion can be differentiated into those non-voluntary forms of behaviour that are not actions (blushing, sweating); the forms of behaviour that are actions, which may be voluntary, involuntary or partly voluntary 21 (smiles, scowls, grimaces, cries, groans, moans, bodily demeanour, expressive gestures, and such actions as jumping for joy or from fright, scratching one's head in puzzlement, punching the air in tnumph); and the forms of behaviour that are modes or manners of acting (tone of voice, manner of gesture). Expressive behaviour, even if it is voluntary, is characteristically noninstrumental. Some forms of expressive behaviour can be exhibited for a purpose, but to that extent its authentic expressive character is diminished. „ i r\ • So, contrary to what Damasio claims, there need be nothJ J hmonotis, contrary to what Damasio ,i • „„„ „„, .„ „„,.., ing hidden about the emotions of others. It is mistaken claims, are commonly perceptible ° to say that ' y ° u cannot observe a feeling in someone else'.-2 It is equally mistaken to think that ' y ° u c a n observe a feeling in yourself'. W e are prone to confuse the fact that we often do not show our feelings, and indeed sometimes make an effort to conceal them, with the misguided idea that the emotions are in some deep sense 'private' and 'hidden'. But this is confused. We can often see delight and rage in a person's face, joy, anguish or horror in their eyes, contempt or amusement in their smile. We can hear the love and tenderness, the grief and sorrow, the anger and contempt, An action is involuntary if it is not done at will, but is of a kind which can be done at will (smiling involuntarily as opposed to voluntarily). An action is partly voluntary if it cannot be initiated at will, but can be suppressed at will (weeping). The matter will be further discussed in chapter 8. Damasio, Feeling of What Happens, p. 42, quoted above, p. 89.

222

Human Faculties and Contemporary Neuroscience

in a person's voice. We can observe the tears of joy or grief, the cries of terror, joy amazement, and the blushes of embarrassment or shame. O n the other hand, to feel emotion oneself— for example, to feel proud or ashamed — is not to observe anything It is important to realize that neither the subjective i Somatic accompaniments of an emo, ,- c , .,./-, the obiective somatic accompaniments 01 an emotion rc tion do not suffice to identify the , ;. . . * ,. , , ., ... themselves are sufficient conditions tor the identificarLilL emotion or to warrant its ascription and ascription of a given emotion. For one's bodily st; subjectively experienced in terms of sensations or objectively determined in physiolog terms, is not an emotion. It is only part of the syndrome of an episode of an emotio perturbation in appropriate circumstances, given the appropriate knowledge, beliefs and concern the agent. Similarly, the behavioural reactions and actions that manifest an emotion do only given the appropriate subjective context of their manifestation. One's hands may tremble v* fatigue, one may sweat because it is hot, and one's throat may be dry because one is thn - and not because one is afraid. One may shed tears (because one is peeling onions), ( may groan {with pain), and one's eyes may be dull (with fatigue) — not because ont grieving. Whether these reactions are manifestations of one emotion or another, or h nothing to do with an emotion, depends upon the circumstances and on what the ag knows or believes of the circumstances in which he finds himself and upon what he a about. , , The connection of emotions with volition is equally conConnectton of emotion and .. „, . _ ., . r ... stitutive of many (but not all) emotions, rear or harm is bound up, other things being equal, with a desire to avoid what is harmful (and the wish or hope that no harm will ensue), remorse with a desire to make amends, love of a person with a desire to protect or further their welfare, shame with the desire to conceal what is shameful. It is important, however, not to confuse a desire with a motive. A desire can furnish a motive. However, a desire for X is not a motive for obtaining X, aithough it may be a motive for doing Y, if doing Y is a means of obtaining X. , Nevertheless, it is no coincidence, but constitutive of Emotion and motive , our concept of emotion that many emotion terms also specify motives for action. We often act out of love or fear, jealousy or compassion. The citation of such motives is one form that explanations of human actions may take. Such explanations of action do not allude to causes, however, but to characteristic patterns of action for the sake of a goal. If one acts from a certain motive, it is not the motive that makes one act, as a man with a gun may make one act by threatening one. Despite the etymology, motives are not ethereal pushes, any more than goals are ethereal pulls. Nor are the emotions that motivate us neural or somatic causes that make us act. Many emotion words also signify motives precisely because emotions typically imply or intimate forms of care and goals for the sake of which an action might be done - for example, the elimination or avoidance of an undesirable state of affairs, the preservation or achievement of a desirable state of affairs. To shake or cry out with fear is not to act with a motive; but to act out of, and be motivated by, fear is (roughly) to act, apprehending one's present situation as threatening, with the goal of eliminating or avoiding the threat. To act out of love is (roughly) to apprehend the object of one's love as lacking a good or as being

Emotion

223

eatened with a harm, and to act for the sake of providing the good or eliminating the m . To act out of gratitude to A is to act with the knowledge that A has previously lefited one, and with the intention of benefiting A because he has previously benefited ; (and if one also acts in order to show one's gratitude, then one further intends that A Cor others) should recognize one's intention). In short, many emotions are also motives for action, not because they are causes of action, but because they indicate a form of concern and structure of belief that informs common patterns of explanation of human action. explaining a human action by saying that it was done out of a motivating emotion, such as fear, love, envy or compassion, is to cite not a factor in an explanation (such as a reason, a cause, a desire, a habit or a tendency), but an explanatory pattern 23

For a more elaborate account of motives, see Kenny, Action, Emotion and the Will, ch. 4, and A. R. "White, The Philosophy of Mind (Random House, New York, 1967), ch. 6.

8

Volition and Voluntary Movement

8.1 Volition The conceptual field subsumed under the categor ,. . -, , ,• i s control . 1 that it involve the exercise of a two-way to do or J power r behaviour thatt is . refrain from doing. Behaviour is voluntary if one can engage in it at will. In this sense, it is behaviour which one can control directly - that is,

226

Human

Faculties and Contemporary

Neuroscience

n o t b y d o i n g s o m e t h i n g else that causes it or stops it (hence n o t as o n e can control one's heartbeat by j u m p i n g up a n d d o w n to increase it or b y lying d o w n to r e d u c e it). A fully voluntary m o v e m e n t is o n e w h i c h the agent controls in its inception, c o n t i n u a t i o n and termination. H e n c e blinking is only partly voluntary, since o n e can blink at will, but c a n n o t control its ' c o n t i n u a t i o n ' or termination, a n d sneezing is only partly voluntary inasmuch as o n e can inhibit it b u t n o t initiate it directly. A voluntary m o v e m e n t is not a m o v e m e n t caused by a Voluntary movement is not move... .-,, , . . r , . ,. . volition or act or will, or by a w a n t , intention or decision r J merit caused by a volition or act of ' ... although w h a t w e d o because w e w a n t to d o it, intend to d o it, or have d e c i d e d to d o it is also s o m e t h i n g w e do voluntarily (unless it is d o n e u n d e r duress or because w e are obliged to d o it b y circumstances). This p o i n t is difficult to grasp, and neuroscientists often suppose, as Descartes and the empiricists (e.g. H o b b e s , Locke, H u m e and B e n t h a m ) did, that voluntary action is m o v e m e n t caused by inner acts of volition. So further explanation is n e e d e d . It is misguided to suppose that voluntary a n d intentional What an act of will and - 1 1 - 1 11 1 J ,„ actions are bodily m o v e m e n t s caused by antecedent acts will-power are _ .„. „, , , -„ r 01 willing to m o v e , i h e r e are such things as acts of will. T h e y are acts performed w i t h great effort to o v e r c o m e one's reluctance or difficulties in acting, typically in adverse circumstances. T h e y are n o t m e n t a l acts called 'willing', which cause bodily m o v e m e n t s . T h e r e is such a t h i n g as w i l l - p o w e r , b u t it is n o t a mental equivalent of muscle p o w e r . R a t h e r , it is d e t e r m i n a t i o n and persistence in pursuit of one's goals in the face of difficulties. T h e r e is such a t h i n g as strength of will, b u t it is not a matter of causally efficacious mental acts of willing, b u t rather a matter of tenacity in sticking to one's purpose. If willing w e r e s o m e m e n t a l h a p p e n i n g that antecedes The incoherent consequences . . . , . . . ., c J of supr .„. , J and is the cause of voluntary action, then, it seems, it would posing willing to be an event or act . have to be either a m e n t a l act or an event. If it w e r e an act, t h e n it w o u l d have to be voluntary. F o r if willing w e r e an involuntary act, t h e n the c o n s e q u e n t behaviour w o u l d n o t b e voluntary either. (If o n e involuntarily knocks over a vase, causing the vase to break, t h e n o n e does n o t break t h e vase voluntarily.) If willing w e r e voluntary, h o w e v e r , it t o o w o u l d have to have b e e n caused by an antecedent volition, for, o n this account, that is w h a t it is for an act of any k i n d to b e voluntary. But that leads to a vicious regress. O n t h e other h a n d , if t h e willing w e r e merely an event that happens w h e n it happens, t h e n the behaviour it causes w o u l d n o t b e voluntarily performed at all — any m o r e than if a feeling caused o n e to sneeze, o n e w o u l d b e sneezing voluntarily. F u r t h e r m o r e , if willing w e r e an antecedent h a p p e n i n g of Why acts of will conceived as , , , , . , . , , • each and every voluntary act, w h i c h it causes, t h e n , given r r causes of voluntary acts are fictions , . , , , -, that w e k n o w w h e t h e r w e acted voluntarily or not, w e should have to b e able to establish the o c c u r r e n c e of such willings (and b e able describe their character and duration). A n d w e should have to b e able to establish that these happenings are the causes of the subsequent b e h a v i o u r . B u t : (a) W e have n o idea w h a t these mythical events of willing m i g h t b e . W e have seen that they c a n n o t b e inner acts. Equally obviously, as w e just n o t e d , they c a n n o t b e feelings

Volition and Voluntary

227

Movement

either. For were willing a feeling, then it would simply be something that happens when it happens, and its causal consequences would not be voluntary actions at all, but movements caused by feelings. (b) We do not know how to identify these acts of willing. But surely, we would have to identify them, to be sure that we did whatever we did voluntarily. And presumably, we niight also misidentify them, and mistakenly think that we did something voluntarily which was actually not caused by any act of willing. Yet we do know (i.e. we can say), without any such identifications, let alone misidentifications, when we act voluntarily, and when involuntarily. (c) We have certainly not identified such inner acts and established a causal relation between them and subsequent bodily movements, on the basis of which we now confidently assert that voluntary acts are acts caused by mental occurrences of willing. Rather, we simply have a picture of what free, voluntary action must be — a picture which rests on neither evidence nor argument. (d) It would surely be absurd to suppose that before each voluntary act there is a separate act of willing. Each of the words of the previous sentence was voluntarily (and intentionally) written down, and each letter in each word was intentionally inscribed. But it is absurd to suppose that in writing them a separate act of willing occurred for each letter and word. (e) It is typically easy enough to identify and distinguish the voluntary and involuntary acts of others. But we do not distinguish these by finding out whether their movements were caused by mental acts of willing, which neither we nor they can identify. When we ascribe responsibility for an action to another person, we do not do so on the basis of identifying an act of will which he has performed and which has caused his bodily movement. O f course, we commonly act because we want to, intend Wanting, intending and deciding are rt

,

.

,

,

., ,

. .

.

,

r

to or have decided to act, either for its own sake or tor

not causes oj actions or movements

the sake of some further goal. But it is mistaken to think that this 'because' is causal. For if it were, then once the want has occurred, the intention been formed, or the decision taken, then we could remain passive, sit back, and let nature take its course. For the action would occur without our taking any initiative. If Í want (have decided, intend) to turn the light on at six o'clock, then when I hear the clock chime 6.00, I should not have to turn the light on in order to fulfil my purpose. I could just let the want (intention or decision) cause my arm to rise and turn the light on. I could say, 'And now it is six o'clock, just look and see my arm rise!' - but that is precisely what does not happen and what one cannot say. Moreover, wants and intentions cannot fulfil the role of acts of will, since they are not acts of any kind. Willing, if it is anything, must be something we do, not a want or a desire that besets us or that we happen to have. And although making a decision can be termed 'a mental act', it is not a cause of behaviour, but a terminus to a state of indecision. Once we have decided, we have formed an intention, and we know what we are going to do. But we have yet to do it — the decision cannot cause us to perform the voluntary action upon which we have resolved. To say that someone did something because he wanted to does not introduce a causal explanation of his action by reference to a mental act or event. But it may serve to exclude

228

Human Faculties and Contemporary Neuroscience

. certain kinds of causal explanation; for example, it excludt Say mo that someone did something , . .r -i • i i , , , involuntary action, so it something was done because th because they wanted to is not to , , .agent wanted to do it, then it was not a mere twitch c & give a causal explanation ' . . an involuntary start. Rather, saying that he did it becaus he wanted to characterizes his behaviour as action, hence as something for which it mak sense to ask for reasons (even though it may not have been done for a reason), as oppose to an involuntary twitch, for which a mere causal explanation is appropriate. Of courst that is perfectly compatible with the existence of a causal explanation of the muscul; contractions involved in his action. There are other possibilities too. One may say, 'I ar leaving the room because I want to, not because you told me to'. This characterizes m action as voluntary and intentional, while excluding one kind of explanation in terms c the specified reason — that is, that you told me to leave. Again, when one acts with further intent, then 'Because I wanted to do so-and-so' or 'Because I wanted to bring th; about' serves to introduce the goal aimed at {e.g. 'I raised my arm because I wanted (t get) the book'). Here one could just as well have said, 'I raised my arm in order to get th book'. Specifying what I wanted gives my reason for acting, not the cause of the move ment of my arm as I raise it voluntarily. If we know what a person wants (aims at) an what his relevant beliefs are, we can often predict his actions, but not because his wants (i conjunction with his beliefs) cause the relevant movements. Indeed, typically we cannc predict how - that is, by what movements — he will execute his intentions, whereas we ca predict his actions. If one knows a chess-player, one can often predict his moves, but wh; enables one to predict his moves will not enable one to predict his movements. 1

8.2 Libet's Theory of Voluntary Movement This fragmentary sketch of a small part of the field c Libet s discoveries ana consequent . . . . ~. ,, , , * volitional concepts suffices to enable us to see what theory awry in a well-known neuroscientific theory of volun tary action. Benjamin Libet has argued that neuroscientific research shows that all volun tary actions are initiated by the brain independendy of any conscious acts of volition. 'Th brain "decides" to initiate or, at least, to prepare to initiate the act before there is an reportable conscious awareness that such a decision has taken place.'" Libet argues ths the neurons in the supplementary motor cortex that are related to a particular physic; movement of a hand start firing 500 ms before the impulses arrive at the muscles involve 1 See J. Gosling, Weakness of the Will (Routledge, London, 1990), p. 183. For more detail on th manifold forms of these familiar non-causal explanations of human behaviour, see A. ]. P. Kerni) Will, Freedom and Power (Blackwell, Oxford, 1975); B. Rundle, Mind in Action (Clarendon Pres: Oxford, 1997); A. R. White, The Philosophy of Mind (Random House, New York, 1967), ch. 6. Fc elaboration of Wittgenstein's ideas on the subject,-see P. M. S. Hacker, Wittgenstein: Mind and Wi (Blackwell, Oxford, 1996), essays VII and VIII. 2 B. Libet, 'Unconscious cerebral initiative and the role of conscious will in voluntary action', i Neurophysiology of Consciousness (Birkháuser, Boston, 1993), p. 276.

Volition and Voluntary Movement

229

in m a k i n g the m o v e m e n t . H o w e v e r , the feeling of intention, w a n t or urge to m o v e the hand, as reported by the subjects, o c c u r r e d only 150 ms before the m o v e m e n t was executed. F r o m this h e concludes (a) that the performance of even a freely voluntary act is initiated unconsciously, some 350 msec before the individual is consciously aware of wanting to move, but also (b) that conscious control of whether the act will actually be performed is still possible during the remaining 150 to 200 msec before activating the muscles. This would appear to preserve the possibility for at least a controlling role for conscious free choice or will. 3 It is i m p o r t a n t to understand that, according to Libet, an act is voluntary if, in addition to 'arising endogenously', w i t h o u t externally i m p o s e d restrictions or compulsions, 'subjects feel introspectively that t h e y are p e r f o r m i n g the act o n their o w n initiative and that they are free to start or n o t to start the act as they wish'. 4 In the e x p e r i m e n t the subjects w e r e 'to choose to perform this act at any t i m e the desire, urge, decision, and will should arise in them. (They w e r e also free not to act o u t any given urge or initial decision to a c t . . . ).' T h e subjects w e r e asked t o m o v e their h a n d if t h e y pleased, and t o n o t e t h e exact t i m e when they felt an urge, desire, wish or i n t e n t i o n to m o v e , w h i c h , as the e x p e r i m e n t showed, succeeded the neural initiation of m o v e m e n t b y 300 ms. T h e premiss of the experiment, as Libet observes, is that this 'subjective event [feeling an urge or desire to move] is only accessible introspectively to the subject h i m s e l f , a n d 'each subject was instructed to " w a t c h for" a n d r e p o r t the earliest appearance of the awareness in question'. 5 , . T h i s e x p e r i m e n t is based o n confused presuppositions. It r Confused presuppositions of .. „ . , . . is neither necessary nor sufficient rfor an act to be volunT,, \, Libet s experiment ' tary that it be p r e c e d e d b y a feeling of desiring, wishing, wanting or i n t e n d i n g to perform it or b y an u r g e to d o it. _ .. ,. ,. heelings or volition are not , , necessary jor voluntary movement

It is n o t necessary, inasmuch as an agent is n o t held, and , , ,, , . ,c , , . . ., ._. does n o t h o l d himself, to have m o v e d involuntarily if h e ' m o v e d w i t h o u t feeling an urge to m o v e or feeling a desire to m o v e . W h e n o n e m o v e s voluntarily - for example, picks u p o n e ' s p e n i n o r d e r to write a n o t e or gets u p in o r d e r to answer t h e doorbell — o n e feels n o urges, desires or intentions, a n d that is n o t because o n e does n o t notice t h e m ! O f course, o n e can say whether one m o v e d voluntarily or involuntarily, b u t n o t o n the grounds that o n e felt an urge, desire or intention just before m o v i n g . ,, . , N e i t h e r is it sufficient for a m o v e m e n t to c o u n t as n heelings of volition are not , . , . , . . , , a- • , r i voluntary, inasmuch as feeling an urge, for example, to b & r stiffiaent jor voluntary movement .3 sneeze j u s t before o n e sneezes does n o t m a k e t h e sneeze voluntary. Indeed, although n o r m a l h u m a n beings can voluntarily inhibit a sneeze, they B. Libet, 'Epilogue: I. Some implications of "time-on theory'", in Neurophysiology of Consciousness, pp. 389f. 5

Libet, 'Unconscious cerebral initiative', p. 270. Ibid., p. 274.

230

Human

Faculties and Contemporary

Neuroscience

c a n n o t sneeze at will at all. Strikingly, Libet's t h e o r y w o u l d i n effect assimilate all hunia voluntary action to the status of inhibited sneezes or sneezes w h i c h o n e did n o t choose t inhibit. For, in his view, all h u m a n m o v e m e n t s are initiated by t h e brain before an awareness of a desire to m o v e , and all that is left for voluntary control is the inhibiting c p e r m i t t i n g of the m o v e m e n t that is already u n d e r way. T h e r e is such a t h i n g as feehng an urge to d o somethinr Movement caused by1 a J felt urge is , , , . , . , , . * r * a n d also as being aware ot a desire to d o something. Bi not voluntary , . ,, . , , ° a m o v e m e n t that is caused by an urge or lelt desire precisely not a voluntary action. If an urge to sneeze, to v o m i t , to cough, etc. causes one t b e h a v e accordingly, t h e n o n e has sneezed, v o m i t e d or c o u g h e d involuntarily. O n e m a y fet an intense desire to drink, eat or m o v e one's h a n d , and, o t h e r things being equal, o n e wi go o n to drink, eat or m o v e one's hand. B u t t h e desire is n o t t h e cause of one's doin so. R a t h e r , o n e drinks in order to assuage one's thirst, eats for a reason - namely, that one h u n g r y - a n d m o v e s one's h a n d on purpose, one's purpose being, for example, to ceas t o u c h i n g s o m e t h i n g repulsive. So t o o , one m a y feel an u r g e to d o s o m e t h i n g , and a( because o n e feels an urge, b u t this 'because' is n o t causal. T h e u r g e o n e feels t o hav a n o t h e r piece of cake does n o t make one's h a n d m o v e irresistibly towards the plate an m o r e than feeling inclined to g o to t h e cinema t o n i g h t will, b y 7 p . m . , cause one's leg to m o v e . If o n e asks one's subjects to m o v e their h a n d voluntaril ., . , . , , , w i t h i n the next m i n u t e , but to take care to n o t e w h e . they feel an urge, an i n t e n t i o n or a desire to m o v e i one's very question subjects t h e m to a t e m p t i n g (but mistaken) philosophical picture c the nature of action and its causal genesis. Indeed, o n e of t h e most interesting (inadvert ent) results of these experiments is that people, w h e n asked to r e p o r t such bizarre things í 'a feeling of i n t e n t i o n to m o v e one's hand', will find such a feehng to report, e v e n thoug it is m o r e than a little doubtful w h e t h e r there is any such t h i n g as 'a feehng of intention Equally, w h e n asked to n o t e w h e n they feel an urge to m o v e , they c o m e up w i t h such feeling, e v e n t h o u g h m o v i n g one's h a n d voluntarily does n o t require and does nc normally involve any such feeling. 6 T h e feeling r e p o r t e d is n o t w h a t makes their move m e n t voluntary, a n d any absence o f feeling w o u l d n o t m a k e it involuntary. T h e fact th; the n e u r o n s in the supplementary m o t o r cortex fire 350 ms before the feeling is allegedl a p p r e h e n d e d does n o t s h o w that the brain 'unconsciously decided' to m o v e before th agent did. It merely shows that the neuronal processes that activate the muscles bega before the time at w h i c h the agent reported a 'feeling of desire' or 'feehng an urge t m o v e ' to have occurred. But, to repeat, a voluntary m o v e m e n t is n o t a m o v e m e n t cause by a felt urge, any m o r e than to refrain voluntarily from m o v i n g is to feel an urge not t m o v e w h i c h prevents o n e from m o v i n g . Libet's auestion presupposed a 1 . , • r misconception of voluntary action

6 This, in our view, has important methodological implications for the design of such experiment What it shows, among other things, is that conceptual confusions on the part of the experiment* are not likely to be weeded out by the subjects upon whom the experiment is made. The latter ai likely to succumb to the very same confusions, or at any rate to be sufficiently intimidated by th experimenter to go along with his forms of description.

Volition and Voluntary Movement

231

We should also remember that a large range of acts are decided on in advance. Reflecting on whether to V this evening, next week or next month, we weigh the reasons for and against F'ing, and decide to V (or not to V). So, when the time approaches (assuming that we have not forgotten and do not change our mind), we V. But to Kthus intentionally, in accordance with our antecedent plans and intentions, could not require that we 'feel an intention' (there being no such thing as a feeling of intention), and does not require that we 'feel a desire'. W e simply act in order to fulfrl our plans, and the relevant movements we make are accordingly voluntary and intentional. , . , Another example of how research on voluntary movement A second examplec of misconceived . . . . " . . . can go awry y due to conceptual contusion is to be round questions in an experiment . „f * in some recent work by C D. Frith and his colleagues. Taking from William James their definition of'willed action' as 'action performed when we consciously pay attention to its selection', they reasoned that 'a deliberate selection is subjectively experienced as willed and occurs when we have a choice of action'. Such 'spontaneous or self-generated actions are not specified by an external trigger stimulus, but are internally driven' by contrast with 'automatic acts where the appropriate response is fully specified by an external stimulus'. Hence, in their view, if a subject is asked to move a finger which is touched by the experimenter, the subject's movement is not 'willed', but is 'automatic'; whereas if the subject is told to move one of two different fingers when one or the other is touched, the movement is 'willed', inasmuch as the subject has 'a choice of action'. To be sure, all such talk of 'willed action' is likely to be confused, and the confusion is bound to be multiplied by James's misconception of voluntary actions and by his incoherent ideo-motor theory." But, disregarding that, it should be obvious that moving a finger that one was requested to move is no less voluntary an act than is moving one or another of two fingers on request.

8.3 Taking Stock „, , , The concepts which we have surveyed in this chapter The concepts surveyed here and in , , . . . ,,, .--,,-,, , , , , ,, , r and the tour previous ones tumi manifold iunctions in r the last jour chapters partly define h u m a n 3ife a n d a r e u s e d t o d r a w a m u l t i t u d e what it is to be a human being ' of fine distinctions. They lie at the heart of our experience, and are indeed partly constitutive of, and define, our experience. For it is because we are concept-using creatures that we can have the wide range of experiences that we do have, many of which are foreclosed to other animals who lack the rich array of concepts available to language-users such as ourselves. For we can know or believe, think or imagine, fear or hope, want or intend, a multitude of things which other animals cannot.

C. D. Frith, K. Friston, P. F. Liddle and R. S. J. Frackowiak, 'Willed action and the prefrontal cortex in man: a study with PET', Proceedings of the Royal Society, B 244 (1991), pp. 241-6. For Wittgenstein's detailed criticisms ofJames's account of volition, see P. M. S. Hacker, Wittgenstein: Mind and Will (Blackwell, Oxford, 1996), pp. 565-8.

232

Human Faculties and Contemporary Neurostience

It should not be surprising to find that the conceptual field is as complex and subtle as the form of life which it informs — our form of life. For the concepts in question are part definitive of what it is to be a human being, and their employment in the expression ai description of our experience is partly constitutive of the life of language-using creatur such as ourselves. The latter point is of paramount importance. The coi They are not theoretical concepts , , ... ., , . , cepts we have been dealing with are not theoretical concq of a science of any kind, although they are rightly invoked and employed in psychoid and in brain neuroscience. 9 Our avowals of love or fear, anger or regret, are not theoreical statements of any kind. Cyrano, pouring out his heart to Rosanne below her balcón was not theorizing about his own behaviour, but expressing his love; Lear, fulminating the ingratitude and perfidy of humanity, was not an amateur theorist conjecturally appling emotion terms to himself in accordance with a popular theory of human nature, h • venting his rage. It is a disastrous confusion, fostered by the eliminative materialists, '' i represent these concepts as part of something called 'folk psychology', which is held to [ a defective, primitive theory of human behaviour. O f course, these concepts are used n< only in the overt expression of emotion, but also in the description of the states of mil and character traits of other people, and in the explanation of human conduct. But it is r equally dire confusion to suppose that all explanation is theoretical. Explanation of hum, behaviour by reference to emotions and motives, knowledge and belief, thought ar imagination, is neither theoretical nor part of a science. The descriptions we have given are only sketches. 'IT The point of our conceptual . . . . i i - >, f conceptual complexity is greater than we have indicad A detailed treatment would require a very long bot • indeed. Nevertheless, our sketches will suffice, we hope, to serve the purpose for which tin are designed. They are meant to remind neuroscientists of the familiar concepts whit they themselves constandy invoke in designing experiments and describing their resuli The sketches of the conceptual articulations involved in the various categories of concerprovide guidelines as to what does and what does not make sense. We hope that neuix scientists will find it useful to consult these sketches (which are, we believe, more acema guides to the concepts involved than those of William James). The examples we h.n examined are intended to alert neuroscientists to the numerous pitfalls that open if insuific ent attention is paid to conceptual clarity. As we remarked at the beginning of chapter the conceptual framework for neuroscience is our familiar psychological conceptual scheni But though familiar, the explicit description of its forms and structure is far from familia and most neuroscientists, in many of their writings, flout them. In flouting them, ilu have not produced a different conceptual scheme, but only incoherences in their employ ment of the existing one. We shall discuss this point in more detail in chapter 14.

9

Obviously one cannot characterize a theoretical concept as a concept that occurs in any stateme of a theory. in We have in mind the conception fostered by Paul and Patricia Churchland in their vaiio writings. We examine these in chapter 13.

Volition and Voluntary

Movement

233

A further p o i n t to w h i c h w e have recurrently drawn attenTlie crvpto-Cartesianism of current . , ^ __ t , - , __ .c , . t l o n 1S t n e e x t e n t • • to w h i c h current neuroscientmc t h o u g h t " is covertly Cartesian. W e have, of course, a c k n o w l e d g e d that c o n t e m p o r a r y neuroscientists, unlike the first t w o generations of t w e n t i e t h - c e n t u r y neuroscientists such as Sherrington, Adrian, Eccles a n d Penfield, are overtly anti-Cartesian. Their anti-Cartesianism consists in their correct rejection of the two-substance dualism of mind (understood as an immaterial substance) and b o d y , interacting, if n o t in the pineal gland, then in the 'liaison brain' or 'the highest brain mechanism'. T h e i r covert Cartesianism, as w e have tried to m a k e clear, consists in the first place in their allocating to the brain a multitude of the psychological functions that dualism allocated to the m i n d . This involves committing the mereological fallacy in neuroscience that was discussed in chapter 3 . It also involves retaining the logical structure or forms of Cartesian explanations of characteristic human psychological functions. W e shall briefly give a recapitulative o v e r v i e w of this point. First, in the case of perception, most neuroscientists accept ' , . . n o t only the distinction b e t w e e n primary a n d secondary r explanation of perception ,. • , , , ,-, ,., -^ , r , qualities adopted by Galileo, Descartes a n d Locke, but also a form of representationalism characteristic of Cartesianism and L o c k e a n empiricism alike (see chapter 4). For they entertain the t h o u g h t that w h a t w e perceive are images or representations of an 'external w o r l d ' that are caused (in the m i n d or the brain) b y t h e stimulation of sense-organs. Secondly, the neuroscientific research o n m e m o r y is Cartesianism in the form of , , ,, , , . , , , j* understandably c o m m i t t e d to the v i e w that m e m o r i e s are ¡ explanation of memory stored in t h e brain. T h e picture underlying this c o n ception is trace t h e o r y that dates back to Aristotle. B u t it is of some interest, in the present context, to n o t e the similarities (and differences) w i t h Cartesian theory. For Descartes thought that the functions that Aristotle and the scholastics ascribed to the 'sensitive soul', including the m e m o r y or 'storage' of sensory impressions or images, can b e fully explained in purely mechanical, corporeal terms. In his Treatise on Man he w r o t e : I should like you to consider . . . all the functions I have ascribed to this machine [the body] - such as the digestion of food, the beating of the heart and arteries. . . the reception by the external sense organs of light, sounds, smells, tastes, heat and other such qualities, the imprinting of the ideas of these qualities in the organ of ' c o m m o n ' sense and the imagination, the retention or stamping of these ideas in the memory . . . I should like you to consider that these functions follow from the mere arrangement of the machine's organs every bit as naturally as the movements of a clock or other automaton follow from the arrangement of its counter-weights and wheels. In order to explain these functions, then, it is not necessary to conceive of this machine as having any vegetative or sensitive soul or other principle of m o v e m e n t and life, apart from its blood and its [animal] spirits. (AT X I , 202, our italics) Quaint physiology apart, there is little in this c o n c e p t i o n of the 'corporeal m e m o r y ' that a contemporary neuroscientist w o u l d wish to gainsay. Disagreement w o u l d break o u t , h o w ever, with respect to Descartes's c o n c e p t i o n of 'intellectual m e m o r y ' , u n i q u e to h u m a n

234

Human

Faculties and Contemporary

Neuroscience

beings, w h o possess a m i n d (or 'rational soul'), a n d w h o accordingly store concepts m tr m i n d . R e j e c t i o n of the Cartesian c o n c e p t i o n of t h e intellectual m e m o r y as lacking ar corporeal foundation is o n e thing. R e t e n t i o n of t h e structural features of t h e Cartesia c o n c e p t i o n of t h e corporeal m e m o r y is another, h o w e v e r , and it is arguably t h e retentio of o n e t h i n g t o o m a n y . For, as w e have argued i n chapter 5, t h e conceptions of memoi storage a n d m e m o r y traces that are here involved are n o t c o h e r e n t . M e m o r y is indee an ability, b u t n o t an ability t o r e p r o d u c e copies of a n t e c e d e n t impressions, images ( ideas. Thirdly, t h e c o n c e p t i o n of mental images as privat . , , , , , inner pictures that only the subject can see is the rt . . ceived c o n c e p t i o n a m o n g n e u r o s a en tists. M e n t a l imagi are c o m m o n l y t h o u g h t to b e copies of antecedent impressions. T h e s e images are sui posed t o be such as can b e scanned, e x a m i n e d and t u r n e d a r o u n d , a n d their features, lil the features of a picture, m a y b e discerned o r o v e r l o o k e d (see chapter 6). Again, th c o n c e p t i o n recapitulates t h e errors of Descartes. In his Conversation with Burman, Descarfc claimed that Cartcsianism in the conception of , . . ; mental images as inner pictures

W h e n external objects act on my senses, they print o n them an idea, or rather a figure, of themselves; and w h e n the mind attends to these images imprinted on the pineal gland in this way, it is said to have sensory perception. "When, on the other hand, the images in the gland are n o t imprinted by external objects, b u t by the mind itself, which fashions and shapes them in the brain in the absence of external objects, w e have imagination. T h e difference between sense-perception and imagination is thus really just this, that in sense-perception the images are imprinted by external objects which are actually present, whilst in imagination the images are imprinted by the m i n d without any external objects. (AT V, 162)

F o r here t o o t h e images of t h e imagination are c o n c e i v e d t o b e just like external picture only 'internal'. This is a part of the Cartesian legacy that, as w e have argued, neuroscien< should jettison. Fourthly, a l t h o u g h , as w e observed i n chapter 7, tl , . a conceptual influence u p o n c o n t e m p o r a r y neur< i. . , scientific investigations of t h e e m o t i o n s is t h e nawc J a m e s - L a n g e theory, it is n o t e w o r t h y that t h e j a m e s i a n c o n c e p t i o n of t h e e m o t i o n s itst is of Cartesian ancestry. As w e have seen, L e D o u x claims that ' t h e brain states a n d bodi responses are t h e fundamental facts of an e m o t i o n ' , t h e conscious feelings being 'addt frills' (see §7.2.1). Damasio holds that an e m o t i o n is 'a collection of changes in a b o d y sta c o n n e c t e d to particular mental images that have activated a specific brain system', and th a feeling of an e m o t i o n is 'the experience of such changes in juxtaposition to the ment images that initiated the cycle' that offer us t h e 'cognition of o u r visceral and musculoskelei state' as it responds to perception o r imagination. In this c o n c e p t i o n they are repeatii some of t h e errors of t h e Cartesian c o n c e p t i o n of t h e passions of the soul as beir 'perceptions, sentiments or e m o t i o n s of t h e soul, w h i c h are referred particularly to t h e so itself, and w h i c h are caused, maintained a n d s t r e n g t h e n e d b y s o m e m o v e m e n t of d Cartesianism in the conception of r J the emotions

rprimary

Volition and Voluntary Movement

235

[niiisi.il] spirits'.11 In general, the passions, according to Descartes, are generated immediately IJV turbulence in the heart, blood and animal spirits12 that produces a mental event - the CN nciienced emotion — in the soul. The proximate physical cause of the feeling of -motion may itself be caused by the perception of a certain (fearful, attractive,, etc.) ^u'Tiial object. Like Descartes, contemporary neuroscientists characteristically fail to dis[ÍIILÍILMI the causes of an emotion from its object. Fifthly, the conception of voluntary action that characCafa-Linism in thejorm of ,,-r t. • • • J u K . , . tenzes current neuroscientiric research is inspired by very r exnhfHitwn of voluntary action , . . / ' much the same picture of voluntariness as informed Cartesian (and British empiricist) reflection. According to Descartes, volitions are actions of CIK1 soul. A subclass of these terminate in our bodies, 'as when our merely willing to walk has the consequence that our legs move and we walk'. 13 'The activity of the soul consists entirely in the fact that simply by willing something it brings it about that the little gland to which it is closely joined moves in the manner required to produce the effect curr^ponding to this volition.' Hence, 'when we want to walk or move our body in some other way, this volition makes the gland drive the spirits to the muscles which serve to hring about this effect'.14 As we have seen above, Libet conceives of a voluntary action as .1 bodily movement caused by an antecedent volition. His purported discovery is that tin* volition itself is an act oí the brain, performed before the human being is aware of any cicin- to move his limbs in accordance with that volition. Here he simply ascribes to the brain the volitional acts which Descartes ascribed to the mind. But the deep misconception that voluntary action is movement caused by an act of volition remains intact. And. as we have seen, it is this that needs to be eradicated. Tlic crypto-Cartesianism of contemporary neuroscientific reflections on the essential psvcliulogical functions of humankind has, we hope, been amply demonstrated in this and tin: previous four chapters. However, it is even more deeply entrenched in contemporary ni'inv^cientific and cognitive-scientific reflections on the nature of consciousness. It is to this theme that we turn in Part III.

1 '«.-.-¡cartes, Passions of the Soul, 1-27. Ihki., 1-46. Ibid., 1-18. frkl., 1-41, 43.

Part III Consciousness and Contemporary Neuroscience: An Analysis

9

Intransitive and Transitive Consciousness 9.1 Consciousness and the Brain Problems concerning consciousness and its nature have , , . .~ . .. ,. . been at the centre of neuroscientiiic, philosophical and r .. . . . , cognitive scientific investigations for the last two decades. a ° Leading neuroscientists have gone so far as to suggest that 'Perhaps the greatest unresolved problem . . . in all of biology, resides in the analysis of consciousness'.1 There are, no doubt, many problems concerning consciousness. Some are empirical problems amenable to scientific investigation. Others are conceptual problems, which can be tackled only by means of conceptual analysis. Distinguishing the two kinds of problem is important, for when a conceptual problem is confused or conflated with an empirical one, it is bound to appear singularly intractable — as indeed it is, for it is intractable to empirical methods of investigation. Equally, when an empirical problem is investigated without adequate conceptual clarity, misconceived questions are bound to be asked, and misguided research is likely to ensue. For, to the extent that the concepts are unclear, to that extent the questions themselves will be unclear. In the following discusión (chapters 9-12), we shall try to shed some light on the concept of consciousness, and to show that clarity concerning the conceptual structures involved in discourse about consciousness has important bearing on the current neuroscientific debate. Neuroscientists are prone to ascribe consciousness to the Ascribing consciousness to the , _ , , , „ , , . „ . , . _ , . . . . , . . , . , . , brain. Gerald Ldelman and Guino iononi claim that brain ¡s a mereological mistake . , . consciousness arises as a particular kind of brain process'; indeed, it is 'a special kind of physical process that arises in the structure and dynamics of certain brains'. 2 Ian Glynn agrees with John Searle (see below) that mental phenomena 'are themselves features of the brain'. 3 Susan Greenfield contends that consciousness 'is an 77;,- importance of distinguishing , . . i , f i-Mh'cptual rjrorn empirical problems , . afout consciousness

T. D. Albright, T. M. Jessel, E. R. Kandel and M. I. Posner, 'Neural science: a century of progress and the mysteries that remain', review supplement to Ceil, 100 (2000) and Neuron, 25 (2000) p. S40. G. M. Edelman and G. Tononi, Consciousness: How Matter Becomes Imagination (Allen Lane, London, 2000), pp. xii and 14. I. Glynn, An Anatomy of Thought (Weidenfeld and Nicolson, London, 1999), p. 396.

240

Consciousness and Contemporary

Neuroscience

e m e r g e n t p r o p e r t y of non-specialized groups of n e u r o n s that are continuously varial w i t h respect to an epicentre'. 4 R u d o l f o Llinás holds that consciousness (or 'mindedness' h e puts it) is a functional state of the brain, o n e ' o f several global physiological compu tional states that the brain can generate', 5 a n d Michael Gazzaniga asserts that consciousn 'is a p r o p e r t y of a neural n e t w o r k ' . 6 M a n y philosophers w i t h an interest in neuroscience c o n c u r . J o h n Searle contends t] ' M y present state of consciousness is a feature of m y brain'. 7 In a similar vein, C o M c G i n n claims that ' T h e brain has some p r o p e r t y w h i c h confers consciousness upon and w o n d e r s w h a t could m a k e the brain ' u n i q u e l y the organ of consciousness'. 8 T c o n c e p t i o n is a particular instance of w h a t w e have called 'the mereological fallacy neuroscience', inasmuch as it involves ascribing to the brain — that is, to a part of animal — an attribute w h i c h it makes sense to ascribe only to the animal as a w h o l e . M a n y scientists, as well as philosophers, argue that at t Neuroscientists and philosophers , 1 1 - 1 1 1 , , ,, , 1 . m o m e n t w e are almost completely m the dark about t f hold that we are largely ignorant 0 / 1 1 1 , , ,, , r . nature of consciousness. Stuart Sutherland, m a mux about the nature oj consciousness q u o t e d remark, w r o t e that 'Consciousness is a fascinati b u t elusive p h e n o m e n o n ; it is impossible to specify w h a t it is, w h a t it does, or w h ) evolved'. 9 C o g n i t i v e scientists, such as Phillip J o h n s o n - L a i r d , aver that ' n o o n e kno w h a t consciousness is, or w h e t h e r it serves any purpose', 1 0 a n d others, such as Da' C h a l m e r s , g o o n to proclaim extravagantly that o u r i g n o r a n c e a b o u t consciousness may 'the largest outstanding obstacle [to] a scientific u n d e r s t a n d i n g of the universe'. 1 1 H o w is this alleged ignorance of the nature of cc The alleged ignorance is commonly . , , . ,-, T-. • , „ , , , . , . , / sciousness to be explained? Daniel D e n n e t t , a lead explained by reference to a mis, . , , * _ conceived notion of privacy philosopher a n d advocate for cognitive science, sugg< that w h i l e science has revealed the secrets of magnetif photosynthesis, digestion and r e p r o d u c t i o n , it has so far failed to penetrate the nature consciousness, for the following reason: particular cases of magnetism or photosynthesis or digestion are in principle equally accessible to any observer with the right apparatus, but any particular case of consciousness

4

S. Greenfield, 'How might the brain generate consciousness', in S. Rose (ed.), From Brains ¡0 Consciousness? (Penguin Books, Harmondsworth, 1998), p. 214. 3 R. Llinás, '"Mindedness" as a functional state of the brain', in C. Blakemore and S. Greenficid (eds), Mindwaves (Blackwell, Oxford, 1987), p. 339. 6 Michael S. Gazzaniga, 'Consciousness and the cerebral hemispheres', repr. in Gazzaniga (ed.), The New Cognitive Neurosciences, 4th edn (MIT Press, Cambridge, MA, Í997), p. 1396. 7 J. R. Searle, Minds, Brains and Science- the 1984 Reith Lectures (BBC Publications, London, 1984), p. 25. 8 C. McGinn, 'Could a machine be conscious?', in Blakemore and Greenfield (eds), Mindwavt's. pp. 281, 285. 9 Stuart Sutherland, Dictionary of Psychology (Macmillan, London, 1989). 10 P. N . Johnson-Laird, The Computer and the Mind (Fontana Press, London, 1988), p. 353. 11 D. J. Chalmers, The Conscious Mind (Oxford University Press, Oxford, 1996), p. xi.

Intransitive and Transitive

Consciousness

241

seems to have a favoured or privileged observer, whose access to the p h e n o m e n o n is entirely unlike, and better than, anyone else's, n o matter what apparatus they have. For this reason and others, not only have w e so far n o good theory of consciousness, w e lack even a clear and uncontroversial pre-theoretical description of the presumed phenomenon. 1 2 Searle too argued that the brain's 'conscious aspects are accessible to m e i n a w a y that they are not accessible to y o u . A n d y o u r present state of consciousness is a feature of y o u r brain and its conscious aspects are accessible to you in a w a y that they are n o t accessible to me.'13 In m u c h the same vein, R i c h a r d Gregory, an e m i n e n t psychologist, suggests that 'Consciousness is difficult to discuss because it is uniquely private: so w e d o n o t have analogies from o u r shared concepts of the physical w o r l d at all adequate to describe o u r 14 A n d w e have already seen (§3.4) that neuroscientists, such as Damasio, e x perience of it'. Edelman and T o n o n i , h o l d that consciousness is an 'entirely private, first-person p h e n o m enon', and that 'privateness' is ' o n e of those fundamental aspects of conscious experience that are c o m m o n to all of its manifestations'. O n this widely shared c o n c e p t i o n , o u r alleged ignorance is explained b y reference to the t h o u g h t that each p e r s o n has privileged access to his o w n consciousness, b u t n o t to the consciousness of others. So consciousness is not a publicly observable, b u t a privately observable, p h e n o m e n o n and, in this respect, unlike the p h e n o m e n a typically studied b y the sciences. For the sciences, it is often held, investígate inter-subjectively verifiable p h e n o m e n a . This alleged difference is held b y s o m e to constitute a m e t h o d o l o g i c a l difficulty. 15 It should be evident from o u r previous discussion i n chapter 3 that this c o n c e p t i o n of privacy is confused. It conflates t h e conceptual truths that there is n o such t h i n g as b e i n g conscious that is n o t a case of someone's b e i n g conscious, a n d that s o m e states o f consciousness can sometimes b e suppressed or concealed, w i t h the conceptual confusions that a person's consciousness is in s o m e deeper sense private and accessible only to the subject, so that only the subject can know truths about it directly. Ignorance is o n e thing, mystery a n o t h e r . N o t only d o Neuroscientists and philosophers hold . . .. , _ , ,, • , . ; ; . scientists a n d philosophers contess to lamentable l g n o r A that consciousness is mysterious * . anee, they also c o m m o n l y allege that consciousness is profoundly mysterious. C r i c k a n d K o c h r e m a r k that consciousness is ' t h e most mysterious aspect' of the m i n d / b r a i n problem. 11 * 'Consciousness', G l y n n observes, 'has always b e e n a

D. Dennett, 'Consciousness', in R. L. Gregory (ed.), The Oxford Companion to the Mind (Oxford University Press, Oxford, 1989), p. 160. Searle, Minds, Brains and Science, p. 25. R. L. Gregory, Mind in Science (Penguin Books, Hannondsworth, 1984), p. 480. Consciousness has a 'subjective' or 'first-person ontology': J. R. Searle, 'The future of philosophy', Philosophical Transactions of the Royal Society, B 354 (1999), p. 2074. 16 F. Crick and C. Koch, 'Mind and brain', Scientific American, 267 (Sept. 1992), p. 111. Interestingly, Crick writes elsewhere of 'our strange feeling of being conscious' (Of Molecules and Men (University of Washington Press, Seattle, 1966)). But is being conscious ¿feeling-: Is being conscious of something or other a feeling? And is there anything strange about it? What would it be like if this 'strange feeling' went away? Would one lose consciousness, i.e. become unconscious'?

242

Consciousness and Contemporary Neuroscience

mystery.'17 Psychologists concur: consciousness, Frisby maintains, 'remains a great my; despite considerable advances in our knowledge of perceptual mechanisms'. 18 And philo sophers and cognitive scientists agree: Dennett holds that consciousness is 'the mos mysterious feature of our minds', 19 and Chalmers asserts that 'Conscious experience is a once the most familiar thing in the world and the most mysterious'. 20 One should be wary when told that something is a deei What a mystery is; the difference _, ,. , ,. . . J , , • , , mystery. There are many subiects about which scienr J between the mysterious and the ' •, ,,. ists are ignorant, and many empirical questions to whicl they do not know the answers. There are some subject about which they are not merely ignorant, but do not have any clear idea, and perhaps no even a vague idea, how to answer the questions that baffle them. These subjects an* questions may be dignified by the name of'mysteries', although, to be sure, one should b careful not to confuse such 'mysteries' - that is, forms of ignorance — with what i wonderful or awesome. For there is much that is wonderful, such as the beauty an fecundity of nature, the impressiveness of great works of art, which is not in the lea; mysterious, and the wonder need have nothing to do with ignorance. Similarly, thj which strikes awe in our hearts, such as great heroism or self-sacrifice, or sublime moun tain peaks and raging storms, is typically not that of which we are ignorant. Not only must one avoid assimilating what is puzzlin The difference between empirical mys, , ra. , . , . , c , . , , ,r . and bathing due to ignorance to what is wonderful c tenes and conceptual mystifications , awesome, but one must also take care not to conrus forms of puzzlement about nature consequent on our ignorance with mystifications con sequent upon conceptual entanglement. One may be too hasty in declaring something to b a 'mystery'. For, in some cases, we do not merely have no clear idea how to discover th truth about a certain subject, we have thoroughly confused ideas. Such confusion ma involve not empirical ignorance and misunderstanding, or inadequate theory and theoret ical understanding, of intractable phenomena, but conceptual confusion. Conceptual con fusion is not the same as error of fact. The latter involves false belief, but the former involves incoherence. It is an error of fact to suppose, as Kepler did, that there are only five planets in the solar system or, as Descartes did, that the pineal gland is the organ in the brain in which signals from the two eyes or the two ears are brought together. But it is a conceptual confusion to suppose that the mind is a kind of entity (either an immaterial substance, as Descartes thought, or the brain, as many suppose today) or that the brain thinks, perceives or calculates (as many neuroscientists suppose), or that self-consciousness is consciousness of a self or an T (see chapter 12). Conceptual confusion consists in transgressing the bounds of sense, and the result of transgressing the bounds of sense is nonsense — that is, a form of words which has no sense. Some kinds of nonsense are patent: for example, 'Is has good' or 'The number 3 fell in love with the number 2 and they got married in world three'. But the forms of nonsense Glynn, Anatomy of Thought, p. 193. j . P. Frisby, Seeing: Illusion, Brain and Mind (Oxford University Press, Oxford, 1980), p. 11. Dennett, 'Consciousness', p. 160. Chalmers, Conscious Mind, p. 3.

Intransitive and Transitive

Consciousness

243

that it is the task of philosophy to disclose are latent nonsense. In particular, the conceptual confusions that bedevil . • i. • j J i_ i_ a us in our reflections on the mind and the brain appear to . ^ make perfectly good sense, it is all too easy to mistake conceptual confusion for empirical ignorance. "When we do so, •we mistakenly suppose that what we need is simply more information and a better theory that will explain the phenomenon that bewilders us. But what we need is more clarity, not so much about the phenomenon as about the concepts we deploy in articulating our lack of understanding. W e are all too prone to project the knots we have inadvertently tied in our own understanding on to the phenomena, and to imagine that the phenomena are singularly mysterious and intractable to human understanding. Some have even declared that the nature of consciousness is a mystery which is in principle beyond the powers of the human mind to fathom. But he is a poor navigator in conceptual waters, who, when he is at sea, declares the land to be unattainable. To think thus is to take the mystification consequent upon our conceptual confusions for an intractable mystery in the nature of things. "We shall argue that the widespread sense of profound It is largely conceptual confusions , . , , c * , , , mystery about the nature of consciousness stems largely about consciousness that generate the ' ' ° ' , from conceptual confusions rather than factual ignorance. r r b sense oj mystery That there are conscious creatures at all is wonderful — and it is meet that we should preserve a sense of wonder at the existence of life in general, and conscious forms of life in particular. There is, to be sure, much that we do not know about the neurological basis of consciousness in its various forms. But the various pronouncements of neuroscientists, scientists and philosophers make it clear that their sense of mystery does not stem from factual ignorance alone, but, as we shall try to demonstrate, from conceptual entanglement. Titat is remediable, by clarification of our concepts and the elimination of the mystification produced by conceptual confusions. And what is left, when clarity has been attained, is ignorance of fact, on the one hand, and a proper sense of wonder at the marvels and contingencies of nature, on the other. To begin at the beginning, we should remind ourselves The subject of human conscious. , , , ,. , ,. , . r . , , , . of some simple conceptual truths which we outlined in ness is the person, not the brain ,. ,. . the preceding discussion. It is not the brain that is conscious or unconscious, but the person whose brain it is. It is not the lecturer's brain that is conscious of anything, but the lecturer, who becomes and is conscious of the ticking of the clock or of the interest or boredom of his audience. The brain is not the organ of consciousness. One sees with one's eyes and hears with one's ears, but one is not conscious with one's brain. There is nothing uniquely private about consciousness. There is nothing private or unobservable about a patient's regaining consciousness after an operation; nor is there anything private about the patient becoming conscious of the nurse tiptoeing about the room - that he has become conscious of her is evident in his smile of recognition, and what he has become conscious of is in full view to any observer. A person who is conscious does not thereby have access to anything, and it is a confusion to suppose that we can uniquely observe our own consciousness. There is nothing inadequate about the concepts we ordinarily deploy to talk about consciousness, although there are indeed abnormal phenomena which lack any common projecting p „a to phenomena and then °' , , , ;,ip that there is a mystery f things

... think- ,, in the

244

Consciousness and Contemporary

Neurosdence

or garden nomenclature and for which psycholoav i • r i J L - i - , round it useful to introduce technical termmoloev (f , « ^ f epileptic automatism). And there is no difficulty of nr ft/ v u i pi ciple in describing the 'presumed phenomenon' of cc sciousness. The phenomenon, or rather phenomena, are not presumed at all — it is nc presumption that human beings and many kinds of animals are conscious creatures commonly conscious of features in their environment, and are subjects of experience - 1 is, suffer pain, perceive things, feel angry, pleased or frightened. N o r is it a presumpt that human beings, unlike mere animals, are self-conscious creatures in the sense that t! have the capacity to reflect on and be conscious of their own mental states, thoughts ; desires, their motives for action, their likings and dislikings, their traits and dispositic and their history. These phenomena are readily describable. But one must be clear u phenomena are correctly characterized as 'phenomena of consciousness'. Finally, in so as the question of what consciousness is for makes sense, there is no difficulty in answer it. But unclarity about our concepts is a seedbed for a misplaced sense of mystery. The role of this chapter is to clarify the ordinary concept of consciousness. Its range application is far narrower than that intended by neuroscientists' and cognitive scienti use of the term 'consciousness' and its cognates. For they are prone to apply the terrr the whole range of human waking experience (and beyond). In the next chapter we s examine some of the reasons for this extension of the terms 'conscious' and 'consciousness'. A first step towards clarity is to distinguish transitive from intransitive consciousness. Transitive consciousness is a matter of being conscious of something or other, or of being conscious that something or other is thus or otherwise. Intransitive consciousness, by contrast, has no object. It is a matter of being conscious or awake, as opposed to being unconscious or asleep. Tfie apparent mysteries of conr , r saousness do not stem from any . , . , inadequacy in our language

r

9.2 Intransitive Consciousness Intransitive consciousness is something that a person or Intransitive consciousness contrasted . , , . r. . , . , . ,, , ., , . animal may Jose (on tainting or being anaesthetized) ana with sleep or unconsciousness , , subsequently recover (when regaining consciousness). ' To be conscious' differs from 'to be awake', if at all, in so far as predicating the latter of a creature implies that it was previously asleep, rather than unconscious. 'Is A awake?' is more appropriate at home, 'Is A conscious?' and 'Has A recovered consciousness?' belong typically in a hospital. To be unconscious differs from being asleep. It may be caused bv fever, anaesthetics, alcohol, etc. It is a state in which a person is not only incapable of perceiving his environment, but also insensible to stimuli, even though parts of his body may react without the person feeling anything. A sleeping person, unlike someone in i coma, dead drunk, or anaesthetized, can be woken by being shaken, by noise, light, ln\it or cold. H e may respond to certain stimuli without awaking — for example, kick oft a heavy blanket as the bed warms up. The range of responses of the sleeping is much greater than that of the unconscious, but in neither case do we attribute any perceptual awareness — the sleeper does not feel hot when he kicks off the blanket, and he does not see the light

Intransitive and Transitive Consciousness

245

wakens h i m , although it is the light stimulus that causes h i m to a w a k e n . The v i r i o n from unconsciousness to sleep, is typically imperceptible (though neurally differen' hie) since it is a transition from absence o f potentialities t o their presence. *"'

B e i n g unconscious a n d b e i n g asleep, b e i n g conscious a n d b e i n g awake, admit o f borderline cases. O r d i n a r y language

rich in n o n - t e c h n i c a l terms for s u c h cases. O n e can b e almost unconscious, s e m i oiiscious, barely conscious, dazed or groggy, half-asleep or n o t fully a w a k e , stupefied, leriumbed, etc. T h e r e are abnormal conditions to w h i c h neither 'conscious' n o r ' t m c o n cious' or 'asleep' apply. S o m e o f these are signified b y familiar terms — for example, delirium', 'hypnotic trance', ' s o m n a m b u l i s m ' - w h i l e others are designated b y technical errns of abnormal psychology — for example, 'fugue', 'epileptic automatism'. p i s c o v e r i n g and explaining the neurological states and processes associated w i t h all these conditions is a task for neuroscience and pharmacology. B u t the fact that p e o p l e and o t h e r diurnal animals are awake or conscious for m o s t o f t h e day, rather than asleep or u n c o n scious, is neither a m a z i n g n o r mysterious. T h e question ' W h a t is consciousness for?', if it were directed at this target, w o u l d patently b e silly. If w e are trying t o p i n p o i n t areas o f scientific ignorance, it is sleep a n d its necessity, rather t h a n b e i n g conscious or a w a k e , that is puzzling a n d d e m a n d s explanation. O n l y of creatures that can be conscious does it m a k e . . . . sense to say that they are ««conscious, just as only creatures that can b e said to b e awake can also be said to be asleep. Only of a living being, in particular a sentient creature, can o n e say that it is conscious or unconscious. H e n c e it is senseless to say this of a m a c h i n e or a tree - a c o m p u t e r that is switched off is n o t unconscious and it does n o t regain consciousness w h e n it is switched on again, and a tree is only figuratively said to a w a k e n in the spring from its winter slumbers. 2 2 C o n s e q u e n t l y , intransitive consciousness and unconsciousness are n o t Tlie subject of intransitive . J

21

Of course, one might say that he 'subliminally' feels hot, sees light, hears sounds, or even that he feels, sees and hears, but is not conscious of doing so or of what he thus feels, hears and sees. But these are merely different (more or less misleading) ways of describing the same phenomena, in which the ordinary criteria for perceiving and feehng sensations are partially satisfied and the ordinary criteria for not perceiving and not feeling sensations are also partially satisfied. 22 Consequendy, Searle's remark (The Mysteries of Consciousness (Granta Books, London, 1997), p. 209): 'I can't prove that this chair is not conscious. If by some miracle all chairs suddenly became conscious, there is no argument that could disprove it' is confused. For there is no such thing as a chair's being either conscious or unconscious. W e have assigned no meaning to the phrase 'a conscious chair', hence too none to the phrase 'an unconscious chair'. T o be sure, one can no more prove that this chair is not conscious than one can prove that the number three is not green or not in love with the number two. But it Searle is suggesting that one cannot prove that this chair is not conscious because of epistemologkal reasons, because one does not have access to another being's consciousness or lack of consciousness, then that is misconceived. For 'This chair is conscious' is a senseless form of words and describes no state of affairs which one might prove to obtain or not to obtain. Moreover, i f ' b y a miracle' (which transgresses the bounds oí sense) all chairs became conscious, as in fairy-tales, one would not need to prove it, one would see it — as the chairs woke from their slumbers, yawned, smiled and started talking to each other. But in such fairy-tales, the chair has a face1.

246

Consciousness and Contemporary

Neurosdence

properties or features of the brain. They are predicable only of sentient creatures, and 1'. other psychological terms, they are predicable only of the creature as a whole, not o| i parts. If neuroscientists speak of the brain and its parts as being conscious, what they • makes sense only in so far as this is a synecdochical or metonymical use of words. The • simply signifies a state or activity of the brain or its parts that is systematically córrela with, and has been found to be a condition of, the creature's being conscious. Whe person who has been under an anaesthetic stirs, groans and opens his eyes, we say that has regained consciousness, that he is awake. But we do not say that his brain is awa For it is not his brain that sits up in bed and asks for a drink, looks around and gets ou bed. The criteria for attributing consciousness to a being - that is, for saying of a creat that it is awake — consist in its behavioural responses to its environment, its perceptual affective reactions and goal-directed actions. In this primary, literal use, it makes no se to attribute consciousness to the brain, for the brain can neither perceive nor fail perceive objects in the environment; only the living animal can do that (althoug1 condition of its so perceiving is that appropriate parts of its brain are functioning norma and its failures to perceive may be due to malfunctioning of those parts of its brain). 1 brain can neither be pleased nor displeased by what is seen, and can neither mani' feelings by expression, gesture or action, nor conceal them by keeping poker-faced. . , , , It has been argued that 'the brain causes consciousnes In what sense the brain is the , . . , . . , , . , llri , ,. .. . We have claimed that it is not the brain that is consci cause oj intransitive consciousness (or unconscious), but the human being whose brain i Can it be said that the brain causes a person to be (intransitively) conscious? It depends what is meant. Of course, if there is such-and-such neural malfunctioning in the brain, •.... person will not be conscious, but unconscious (or suffering from akinetic mutism, epileptic fugue, etc.). Equally, when a person awakens or regains consciousness, the intralarninar nuclei trigger cortical activity, but for which he would not awaken or regain consciousness. On the other hand, the answer to the question 'What is the cause of A's awakening?" is not 'Why, his brain, of course'. Rather, the brain events, triggered by the intralanuuar nuclei, are viewed as causal conditions of awakening and/or of being awake, the 'precipitating cause' being such an event as a loud noise, being shaken by the shoulder, etc. (This is parallel to the fact that we do not normally identify the cause of a fire with the presence of oxygen in the air, even though in the absence of oxygen there would be no fire.) _, ,. There is nothing essentially private about intransitive con1 here is nothing intrinsically pnv^, , , . , , . ¿ •. • sciousness. Inat another person has regained consciou.iate about intransitive consciousness ., ° ness or awoken is normally fully visible in his behaviour. "We can ordinarily see that a person is conscious (also what he is conscious of and ven, often also what state of consciousness he is in — of which more below). It is true th.it someone may pretend to be unconscious, and take us in — but no one can pretend to be conscious. People may dissimulate and veil their feelings or conceal their current state oi consciousness — that is, what state of mind they are in. But what can be thus suppressed or concealed is also what can be manifest or revealed. It may not be evident whether

Searle, Minds, Brains and Science, pp. 18-22.

Intransitive and Transitive

i e o n e Was

Consciousness

247

or was n o t conscious of s o m e t h i n g , just as it m a y n o t be evident w h e t h e r

leone noticed s o m e t h i n g . B u t it also m a y b e obvious that h e did — w h o l l y evident in changes in his behaviour. O f course, that I a m conscious is n o t s o m e t h i n g that is o n i-^g grounds of any behavioural criteria.

ThefiK -y

evj¿en]. to m e

But nor is it evident b y introspection or privileged access of any kind. If it seems as if any «articular case of consciousness has 'a favoured or privileged observer, w h o s e access to the phenomenon is entirely unlike, a n d better than, a n y o n e else's', 24 then it w r o n g l y seems so. That one is conscious is n o t a piece o f information w h i c h o n e m i g h t lack a n d acquire by having access to it by some means or other. I m a y b e c o m e , a n d so be, conscious of y o u r regaining consciousness, b u t I c a n n o t b e c o m e , and be, conscious of m y regaining c o n sciousness. M y o w n (intransitive) consciousness is n o t an object of possible experience for me, but a p r e c o n d i t i o n for any experience. 2 5 **ie f

J }

C o n s e q u e n t l y , there are n o t , a n d could n o t be, any grounds or evidence for claiming to b e conscious. T o say

'I a m conscious' is n o t to m a k e a claim, and there can b e no grounds for saying this. O n e c a n n o t mistakenly or falsely claim that o n e is conscious, as one may mistakenly or falsely claim that a n o t h e r person is conscious. If s o m e o n e said in his sleep 'I a m conscious', w e w o u l d n o t accuse h i m o f m a k i n g a false claim, any m o r e than if h e said 'I a m asleep', w e w o u l d say ' H e is quite right'. 2 6 It c a n n o t seem to o n e that one is conscious. F o r o n e c a n n o t say, 'It seems to m e that I a m conscious, b u t I m a y b e wrong', or 'It seems to h i m that h e is conscious, so h e is probably right'. H e n c e it is misguided to claim that 'If it consciously seems to m e that I am conscious, t h e n I a m conscious', 2 7 and e v e n m o r e confused to try to describe 'those features of m y life and the nature of my acquaintance with them that I w o u l d cite as m y " g r o u n d s " for claiming that I am — a n d d o n o t merely seem to be — conscious'. 2 8 T h e r e is a use for the sentence 'I am conscious', b u t it is n o t to express an i t e m of indubitable, privileged k n o w l e d g e or to convey to others one's private observations or to report one's current experience. It is rather akin to a signal. As I recover consciousness after an anaesthetic, I m i g h t say to a nurse, w h o m I notice t i p t o e i n g a r o u n d the r o o m , 'I a m conscious'. I d o n o t say this after 'observing m y o w n consciousness', b u t after observing that she thinks I a m still

24

Dennett, 'Consciousness', p.160. ~3 It might be objected that since one dreams when one is asieep, one can have experiences even though one is not conscious. But to dream is not to have any experiences, although it may involve dreaming that one has experiences (if one dreams about oneself, and dreams that one perceives, does and undergoes various things). However, to dream that one is watching a football match, enjoying a party, etc. is not to have any experience at all - it is only to dream that one is having such experiences. 2( ' L. Wittgenstein, Zettel, ed. G. E. M. Anscombe and G. H. von Wright, tr. G. E. M. Anscombe (Bkckwell, Oxford, 1967), §396. Searle, Mysteries of Consciousness, p. 122. D. Dennett, 'Towards a cognitive theory of consciousness', repr. in his Brainstorms (Harvester Press, Brighton, 1981), p. 173.

248

Consciousness and Contemporary

Neuroscience

unconscious. I might just as well say ' H e l l o ' or ask w h a t t i m e it is. 2y F o r these utter n o less than 'I a m conscious', s h o w that I have regained consciousness. T h e latter, howevp also asserts w h a t its utterance shows.

9.3 Transitive Consciousness and its Forms Transitive consciousness m a y b e dispositional or occm Dispositional and occurrent transitive ™ ,_,., _ 1 1 ... ., , , . rent. W h e n w e say 01 a person that h e is conscious rt consciousness distinguished; being . e 1 1 • rj. his ignorance or expertise, or conscious of his superior r B r conscious oj and being aware of dis' ¿"j^uor c tinguished inferior social status,we are typically speaking of a dtspos tion or tendency h e has to b e conscious, from occasion t occasion, of these things. H e n c e , t o o , w h e n w e speak of a person as b e i n g dass-consciou m o n e y - c o n s c i o u s or safety-conscious, w e are indicating a proneness to b e conscious of h or others' social background, of financial considerations, or o f matters pertaining to safety Occurrent consciousness, by contrast, is a m a t t e r of currently b e i n g conscious of somethin or conscious that s o m e t h i n g is thus-and-so. 3 1 "We cannot, i n this sense, b e conscious c m a n y things at the same time, since w e c a n n o t have o u r attention held by m a n y t h i n g s ; o n c e or have o u r m i n d occupied by m a n y different things at t h e same time. N o r can w remain conscious of w h a t n o longer holds o u r attention or occupies o u r thoughts. In th respect, being conscious of s o m e t h i n g differs from being aware of something. For where; o n e ceases to b e conscious of s o m e t h i n g that ceases to engage o n e ' s attention or to occup one's thoughts, o n e remains aware of things o f w h i c h o n e has b e e n informed, as Ion as o n e does n o t forget t h e information i n question a n d adverts t o it reasonably frequent! 1 So w h a t e v e r o n e is conscious of, o n e is also aware of, b u t o n e m a y b e aware of thinj of w h i c h o n e is n o t conscious. T h e following discussion is c o n c e r n e d primarily wit o c c u r r e n t transitive consciousness. Intransitive consciousness is a condition for the various forms of o c c u r r e n t transitive consciousness — that is, for being conscious of s o m e t h i n g at a given time, for a time. Borderline cases of intransitive consciousness yield borderline cases of transitive consciousness. O n e m a y be vaguely or half-conscious of s o m e t h i n g w h e n one is half-asleep or semi-conscious.

29 L. Wittgenstein, Philosophical Investigations, ed. G. E. M. Anscombe and R. Rhees, tr. G. E. M. Anscombe (Blackwell, Oxford, 1953), §416. 30 The following discussion is much indebted to A. R. White, Attention (Blackwell, Oxford, 1964), ch. 4. 31 To be conscious of something is not, we note in passing, a mere matter of 'having a thought about it or a sensation of it' (D. M. Rosenthal, 'Thinking that one thinks', in M. Davies and G. W . Humphreys (eds), Consciousness (Blackwell, Oxford, 1993), p. 198). One does not, pai . rt , . , between occurrent and dispositional mental statps « i saous ana unconscious mental states . . ' *• the distinction between mental states such as being elate excited, cheerful, chronically depressed or anxious, on the one hand, and items which a neither occurrent nor dispositional mental states, such as knowing, believing, opinin intending, on the other, is a confusion. What is needed, it might be said, is the distincti( • between conscious mental states and mental states that are unconscious. So, for examp! Searle contends that 'the belief that the Eiffel Tower is in Paris is a genuine mental stai even though it happens to be a mental state that most of the time is not present consciousness'.11 A conscious belief state, therefore, is manifest when one's belief is 'prese i to consciousness'. So almost all one's beliefs are normally unconscious, no matter wheth one is awake or asleep. They are only conscious if one is currently thinking or 'occurrenbelieving', for example, that the Eiffel Tower is in Paris. So one is literally in uncountal many different unconscious mental states at any given time, as many as the uncountal many different beliefs (memories, intentions, items of knowledge) one has.12 But this misconceived, not only because believing is not a mental state of any kind, but a) because the concept of an unconscious belief is misconstrued. . r„ . , We must distinguish between believing and what is b J lieved, both of which are confusingly called 'beliefs' (§6. If a person has a passionate or tentative belief that p, then he passionately or tentatrv» believes that p, but what he believes (also called 'a belief), namely that p, is neither passion, nor tentative. O n the other hand, if his belief is certain, doubtful, probable or possib then it is certain, doubtful, probable or possible that p, but his believing what he belies need be neither certain or doubtful, nor probable or possible. We must be careful not confuse the ascription of predicates to a person's believing with the ascription of predica to what he believes. Bearing this distinction in mind, what is supposed to • , a mental state? Clearly the believing, not what is believi since the belief that the Eiffel Tower is in Paris is obviou no mental state, even if the believing might (wrongly) be thought to be. Now, what supposed to be 'present to consciousness' when a belief is conscious, and unconsciou1 that is, 'not present to consciousness' — when it is not? Evidently it is not the believing, for then what one would be conscious of would be that one believes whatever one believes,

11

J. R. Searle, The Rediscovery of the Mind, (MIT Press, Cambridge, MA, 1992), p. 154. Uncountably many, not merely because there are practical difficulties in counting a person's beliefs, but because there are no clear principle's, of countability. When [ beheve that the roses in your garden are red, is that one belief or many? For do I not also believe that the roses in yo;>r garden are not green, not blue, not yellow or orange, etc.? And when I believe that the carpet is 1.Í feet long, do I not also believe that it is not 10 or 11 and not 13, or 14, etc. ad infinitum, feet long? 12

Conscious Experience,

Mental States and Qualia

269

what one believes - for example, that the Eiffel Tower is in Paris. But, according to rle the belief that p is conscious when the thought that p crosses one's mind, or when e bears it in mind that p. Although there is here an important distinction, it is not -ween a conscious and an unconscious belief, but the distinction between bearing in nd or thinking about something one believes to be thus-and-so and not bearing it in rtd or thinking about it. I believe that Hannibal should have besieged R o m e after rinae, that no one poet wrote both the Iliad and the Odyssey, and myriad other things I rarely think about most of what I believe, but that does not make it unconscious, any >re than my knowledge that the Battle of Bannockburn was fought in 1314, Agincourt 1415, and my knowledge of a myriad other things too, is unconscious knowledge. An unconscious belief, correctly construed, is something couscw J Y j believe, believing which colours my emotional reacs tions and informs my actions, but which I am unwilling to nowledge, either to myself or to others, as something which I believe. Nevertheless, that 3 believe explains my aberrant behaviour (my neuroses) — or so it is conjectured. Only in special circumstances - for example, in the course of psychotherapy - do I come to recognize that I unconsciously believed this or that about myself, my childhood or my parents. And my recognition, under analysis, that I so believe is held to be confirmation of the explanatory hypothesis. It is this that is called 'an unconscious belief. But there is nothing unconscious about my myriad beliefs (i.e. what I believe), about which I am not thinking. Nor is any one of the myriad things I know — for example, that Hastings was fought in 1066 - 'unconscious knowledge' just because I am not thinking about that known fact, any more than my intention to take a holiday in Naples this summer is an unconscious intention just because 1 do not continuously think about what I so intend. Furthermore, while I may now believe that p and later A confusion about 'occurrently , ,. , . ,,, - , ,. , . , * cease to believe that p, it would be misleading to speak of my occurrently believing something as opposed to my currently believing something. To believe is not an act or activity that one may be engaged in, and there is no answer to the question 'What are you doing?' which runs 'I am believing that p\ Nor is there a well-formed question 'What are you now believing?' (as opposed to 'What do you now believe?'). The fact that 1 am now thinking about Hannibal's invasion of Italy, and reflecting that he should have besieged Rome after Cannae, does not imply that I am now 'consciously believing' that he should have done so, whereas normally I am in an unconscious state of so believing. To think about something which 1 believe is not to believe it 'consciously', let alone to 'be believing' it 'consciously'; and not to think about something I believe is not to believe it or 'be believing' it 'unconsciously'. ,,. . , , , Finally, while the concept of a mental state is indeed Misusing the term mental state , , . , , , vague and elastic, there is no reason to stretch it to the point at which it becomes legitimate to talk of being in tens of thousands of (unconscious) mental states at the same time. There is nothing to be gained by adopting this novel form of speech, and our common notion of a mental state is thereby blurred and rendered obscure to no purpose. In the licit use of the expression, one can no more be in, suffer or

270

Consciousness and Contemporary Neuroscience

enjoy indefinitely many mental states at a given time than one can attend to, or be occupied with, tens of thousands of things at a given time. There is another misuse of the term 'unconscious' that i< Activities Jof the brain are neither i n j . . ™ " . . common among psychologists and neuroscientists. The\i } unconscious nor conscious " are prone to talk of the activities of the brain as being 'unconscious'. So, for example, it is said that 'much of the brain's work is done unconsciously, by innumerable small bits of specialized brain tissue', and that 'the great bulk oi these tissues are unconscious'; that when reading the phrase 'visual focus', we are noi conscious that the word 'focus' in this phrase is a noun, but the phrase would be incomprehensible if the brain 'did not label "focus" as a noun unconsciously'; and similarly, that when we read or, more generally, engage in highly practised 'automatized' skills, 'we do much more work unconsciously than consciously'.13 But this is confused. The tissues of the brain, like the brain itself, are neither conscious nor unconscious, neither awake nor asleep. The 'work' done by the brain is not done consciously and deliberately, but nor is it done unconsciously and without deliberation. For the brain cannot do anything consciously or unconsciously, since it is not a conscious creature with the capacity to be conscious or to be conscious of anything, let alone with the capacity ol doing anything either with or without deliberation and attention. Rather, unless we are in a laboratory under a visible scan, we cannot perceive activities of our brain, any more than we can perceive the back of our heads or the far side of the moon. And since normally we cannot perceive the activities of our own brain, we cannot become and then be conscious of them; but this does not mean that they are unconscious activities, either of ourselves 01 of our brains. The fact that we are not conscious of the activities of our brains does nol imply that we are engaged in them unconsciously; but nor does it imply that the brain i; engaged in these activities unconsciously. For only a creature that can do something consciously can also be said to do something unconsciously. It is true that when we read the phrase 'visual focus', we do not reflect on the fact thai it is a noun — we take it for granted that it is, and read it as a noun. This does not mean that our brain unconsciously labels 'focus' as a noun, for the brain does not label anything, either as a noun or as anything else. We know that 'focus', in this occurrence, is a noun; but knowing is not an activity we engage in. We do not have to parse the phrase in order to understand it. But this does not mean that we parse it 'unconsciously'. More generally, many of our skills can be practised without thought, but this does not mean that the thought that would be necessary for a beginner now goes on 'unconsciously'; it means that it no longer goes on at all. It does not have to go on, precisely because we have now acquired the relevant skill. (Someone who has learnt to touch-type no longer needs to look at the keyboard, but this does not imply that he looks at it unconsciously or that he has to work out where the keys are unconsciously.) A small child may need to spell out each letter as he laboriously reads, an adult can scan a sentence at a time. The adult has an ability the child lacks: namely, to take in a sentence as a whole. He would presumably not have that ability unless certain synaptic connections in the brain were in place and certain

Baars, Theater of Consciousness, pp. 4, 6, 17.

Conscious Experience, Mental States and Qualia

271

brain events were taking place when the ability is exercised. But this does not imply that either he or his brain spells out each letter or word unconsciously.

10.3 Qualia The temptation to extend the concept of consciousness Qualia conceived of as the qualitat, . . , ^ , . , to encompass the whole domain of experience was r r ¡ve character of experience -the , , . . , . philosophers' conception &*&? strengthened by philosophers misconceived mtro" duction of the notion of qualia. Unfortunately, neuroscientists picked up this aberrant idea and the misconceptions associated with it. The term 'qualia' was introduced to signify the alleged 'qualitative character of experience'. Every experience, it is claimed, has a distinctive qualitative character. Qualia, Ned Block holds, 'include the ways it feels to see, hear and smell, the way it feels to have a pain; more generally, what it's like to have mental states. Qualia are experiential properties of sensations, feelings, perceptions and . . . thoughts and desires as well.' 14 Similarly, Searle argues that 'Every conscious state has a certain qualitative feel to it, and you can see this if you consider examples. The experience of tasting beer is very different from hearing Beethoven's Ninth Symphony, and both of those have a different qualitative character from smelling a rose or seeing a sunset. These examples illustrate the different qualitative features of conscious experiences.'15 Like Block, Searle holds that thinking has a special qualitative feel to it: 'There is something it is like to think that two plus two equals four. There is no way to describe it except by saying that it is the character of thinking consciously "two plus two equals four".' 16 The subject matter of an investigation of consciousness, Chalmers suggests, 'is best characterized as "the subjective quality of experience"'. A mental state is conscious, he claims, 'if it has a qualitative fed — an associated quality of experience. These qualitative feels are also known as phenomenal qualities, or qualia for short. The problem of explaining these phenomenal qualities is just the problem of explaining consciousness.'17 He too takes the view that thinking is an experience with a qualitative content: 'When I think of a lion, for instance, there seems to be a whifF of leonine quality to my phenomenology: what it is like to think of a lion is subtly different from what it is like to think of the Eiffel tower.' 18 .„ Neuroscienhsts follow the ,., , philosophers

Neuroscientists have gone along with the notion of qualia. ^,, , , ,.., , ,. . . Ian Glynn contends that Although qualia are most obviCT l

ously associated with sensations and perceptions, they are also found in other mental states, such as beliefs, desires, hopes, and fears, during conscious episodes of these states'.19 Damasio states that 'Qualia are the simple sensory qualities to be Ned Block, 'Qualia', in S. Guttenplan (ed,), Blackwel! Companion to the Philosophy of Mind (Blackwell, Oxford, 1994), p. 514. J. R. Searle, 'Consciousness', Annua! Reviews, 23(2000), p. 560. 16 Ibid., p. 561. Chalmers, Conscious Mind, p. 4. m Ibid., p. 10. I. Glynn, An Anatomy of Thought (Weidenfeld and Nicolson, London, Í999), p. 392.

272

Consciousness and Contemporary

Neuroscience

found in the blueness of the sky or the t o n e of a s o u n d p r o d u c e d by a cello, a n d the fundamental c o m p o n e n t s of the images [of w h i c h p e r c e p t i o n allegedly consists] are thus m a d e up of qualia'. 2 0 E d e l m a n a n d T o n o n i h o l d that 'each differentiable conscious experience represents a different quale, w h e t h e r it is primarily a sensation, an image, a thought, or even a m o o d 1 , a n d go o n to claim that ' t h e p r o b l e m of qualia 7 is 'perhaps the most d a u n t i n g p r o b l e m of consciousness'. T h e subjective or qualitative feel of a conscious experi. . , ,_ - , • i • c ¿\ ence is in t u r n characterized m terms oi there being , . , . , . , , • i i something it is like for an organism to have t h e experience. A ° W h a t it is like is the subjective character of the experience. ' A n experience or other mental entity is " p h e n o m e n a l l y c o n s c i o u s " ' , t h e Routledge Encyclopaedia of Philosophy tells us, 'just in case there is s o m e t h i n g it is like for o n e to have it.' 2 2 ' C o n s c i o u s states are qualitative', Searle explains, 'in the sense that for any conscious state . . . there is s o m e t h i n g that it qualitatively feels like to b e in that state.' 2 3 T h e idea, a n d the mesmerizing turn o f phrase 'there is s o m e t h i n g w h i c h it is like', derive from a paper b y t h e philosopher T h o m a s N a g e l entitled ' W h a t is it like to b e a bat?'. Nagel argued that 'the fact that an organism has conscious experience at all means, basically, that there is something it is like to be that organism. . . . fundamentally an organism has conscious mental states if and only if there is s o m e t h i n g it is like to be that organism - s o m e t h i n g it is like for t h e organism'. 2 4 This - that is, w h a t it is like for t h e organism - is the subjective character or quality of experience. If w e take for granted that w e understand the phrase Nagel's explanation of consciousness , , . ,, - . , • . ,., , t1_ .u A & r i , , there is s o m e t h i n g w h i c h it is like thus used, t h e n it in terms of there being something it ,, . . . , „ . .... seems that N a g e l s idea gives us a handle o n the concept of a conscious creature a n d o n the c o n c e p t of a c o n scious experience: (\) A creature is conscious or has conscious experience if and only if there is something which it is like for the creature to be the creature it is. (2) An experience is a conscious experience if and only if there is something which it is like for the subject of the experience to have it. So, there is s o m e t h i n g w h i c h it is like for a bat to b e a bat (although, N a g e l claims, w e c a n n o t imagine what it is like), a n d there is s o m e t h i n g w h i c h it is like for us t o b e h u m a n beings (and, h e claims, w e all k n o w w h a t it is like for us to b e us). Explaining the qualitative character r . . , , . r of experience in terms of there bans \,. .. . ... t , -t something it is like to have it

20

A. Damasio, The Feeling of What Happens (Heinemann, London, 1999), p. 9. Note that there is here an unargued assumption that colour and sound are not properties of objects but of sense impressions. 21 G. M. Edelman and G. Tononi, Consciousness: How Matter Becomes Imagination (Allen Lane, The Penguin Press, London, 2000), p. 157. 22 £. Lomand, 'Consciousness', in Routledge Encyclopaedia of Philosophy (Routledge, London, 1998), vol. 2, p. 581. 21 Searle, Mystery of Consciousness, p. xiv. 24 T. Nagel, 'What is it like to be a bat?', repr. in his Mortal Questions (Cambridge University Press, Cambridge, 1979), p. 166.

Conscious Experience, Mental States and Qualia

273

It is important to note that the phrase 'there is something which it is like for a subject to have experience E' does not indicate a comparison. Nagel does not claim that to have a given conscious experience resembles something (e.g. some other experience), but rather that there is something which it is like for the subject to have it; that is, 'what it is like' is intended to signify 'how it is for the subject himself .^ It is striking, however, that Nagel never tells us, with regard to even one experience, what it is like for anyone to have it. He claims that the qualitative character of the experiences of other species may be beyond our ability to conceive. Indeed, the same may be true of the experiences of other human beings. 'The subjective character of the experience of a person deaf and blind from birth is not accessible to me, for example, nor is mine to him.' But we know what it is like to be us, 'and while we do not possess the vocabulary to describe it adequately, its subjective character is highly specific, and in some respects describable in terms that can be understood only by creatures like us'. 26 Philosophers and neuroscientists have gone along with , . ., , , , ., this idea. It seems to them to capture the essential nature of conscious beings and conscious experience. Thus Davies and Humphreys contend that, 'while there is nothing that it is like to be a brick, or an ink-jet printer, there is, presumably, something it is like to be a bat, or a dolphin, and there is certainly something it is like to be a human being. A system - whether a creature or artefact — is conscious just in case there is something it is like to be that system.'27 Edelman and Tononi agree that 'We know what it is like to be us, but we would like to explain why we are conscious at all, why there is "something" it is like to be us — to explain how subjective experiential qualities are generated'. 28 And Glynn holds that with respect to our experiences — for example, of smelling freshly ground coffee, hearing an oboe playing, or seeing the blue of the sky — 'we know what it is like to have these experiences only by having them or by having had them. . . .Just as it feels like something to smell freshly ground coffee, so it can feel like something (at least intermittently) to believe that. . . , or to desire t h a t . . . . or to fear that. . . ,'29 Qualia, then, are conceived to be the qualitative characteristics of 'mental states' or of 'experiences', the latter pair of categories being construed to include not only perception, sensation and affection, but also desire, thought and belief. For every 'conscious experience' or 'conscious mental state', there is something which it is like for the subject to have it or to be in it. This something is a quale — a 'qualitative feel'. 'The problem of explaining these phenomenal qualities', Chalmers declares, 'is just the problem of explaining consciousness.'30 Philosophers and neuroscientists

26

Ibid., p. 170n. Ibid., p. 170. Davies and Humphreys (eds). Consciousness, p. 9. Edelman and Tononi, Consciousness, p. 11. Glynn, Anatomy of Thought, p. 392. Chalmers, Conscious Mind, p. 4.

274

Consciousness and Contemporary Neuroscience

10.3.1 'How it feels' to have an experience One reason given for extending the concept of conThe primary rationale for extending . , , . , .. . u , . , ,. ' i . sciousness beyond its legitimate conservative boundaries the ordinary concept of consciousness ' , . , 1 1 1 - 1 1 was that what is distinctive, remarkable, indeed mysterious, about experiences is that there is something which it is like to have them. An experience, it is argued, is a conscious experience just in case there is something which it is like for the subject of the experience to have it. Consciousness, thus conceived, is defined in terms of the qualitative feel of experience. There is a specific way it feels to see, hear and smell, to have a pain, or indeed 'to have mental states' (Block); every conscious state has a certain qualitative feel to it (Searle), and each differentiable conscious experience represents a different quale (Edelman and Tononi). This qualitative feel, unique to every distinguishable experience, is what it is like for the subject of the experience to have the experience. Or so it is held. Our suspicions should be aroused by the odd phrases used to invoke something with which we are all supposed to be utterly familiar. W e shall examine 'ways of feeling' first, and there being 'something which it is like' subsequently. Is there really a specific way it feels to see, hear or smell? Is there always a way it feels to ~ - I - . J I 1 i_ i_ J u- - i , , . . ., One might indeed ask a person who has had his sight, have a conscious experience ? _ .. , ' ' same objection of illegitimate reiteration. The interpolated phrase 'for a human being' cannot play the role which a phrase in that position is meant to play. But there is perhaps another source of unease. Proteus, avatars and gods apart, a human being (unlike a soldier or a sailor who can abandon his vocation) cannot cease to be a human being without ceasing to exist. Nor can anything other than a human being be a human being. So neither principle of contrast is satisfied. One cannot ask 'What is it Hke for a human being,

280

Consciousness and Contemporary Neurosdence

as opposed to another creature, to be a human being?', for, mythology apart, nothing other than a human being can be a human being. Nor can one ask 'What is it like for j human being to be a human being, as opposed to being something else?', since there is no other creature a human being might be. (Similarly, there is nothing other than a bat which might be a bat, and there is no other creature a bat might be, other than a bat.) So, if any sense can be made of the question 'What is it like for a human being to be a human being?', it collapses into the question 'What is it like to be a human being?*. This question is curious. One might take it to mean: 'What is human life like?' That is a nebulous question indeed. It might be answered variously: for example, 'Nasty, brutish and short' or 'Full of hope and fear'. Similarly, 'What is it like to be a bat?', if it means anything, can be no more than a request to describe the hfe of a bat in a comparable manner. There seems no difficulty of principle in doing that, but there is no reason to suppose that it sheds any light on the nature of consciousness. But it is true that the question can be asked only of conscious creatures who take pleasure in certain things, fear other things, find interest in things, and so forth. Similar arguments apply to the claim that there is something which it is like for us to be us or for me to be me, and that we all know what it is like. It makes no sense to ask what it is like for me to be me, for no one else could be me, and I could be no one other than myself 'I am me', apart from being ungrammatical, says nothing (what could it be intended to say — to affirm an instance of the identity of a thing with itself?).33 So 'There is something which it is like for me to be me' likewise says nothing. Not only do I not know what it is like for me to be me, there is nothing to know. The claim that there is something it is like for me to be a human being is equally questionable. For the question 'What is it like for you to be a human being?' presupposes that I might be, or have been, something other than a human being, something that might be contrasted with my being a human being - and there is no such thing. So, let us take stock: Incoherences that ensue (i) The sentences 'There is something which it is like to be a human being', 'There is something which it is like to be a bat', and 'There is something which it is like to be me', as presented by the protagonists in this case, are one and all awry. (ii) The question 'What is it like for an X to be an X?' is illicit because of the reiterated term, and, if 'X' is a name of a kind of animal (like 'human being' or 'bat'), doubly at fault. The most that can be made of it is to interpret it as equivalent to 'What is it like to be an X?', and to interpret that question as an inquiry into the characteristic attitudinal features of the life of an X. Such questions can be answered, and one need not be an X or similar to an X in order to answer them. O n e merely has to be weD informed about the lives of Xs. (iii) The questions 'What is it like for me to be me?' and 'What is it like for me to be a human being?' are equally illicit.

33

When Shakespeare's Richard III mutters with relief'Richard is himself again', he does not mean that he had not been identical with himself but that now, mercifully, he is again!

Conscious Experience, Mental States and Qualia

281

If this is correct, then it is wrong for Nagel to suggest that 'we know what it is like [for | u s to be us', that there is something 'precise that it is like [for us] to be us', and that 'while we do not possess the vocabulary to describe it adequately, its subjective character is highly specific'. It is mistaken of Edelman and Tononi to assert that we all 'know what it is like to be us', and confused of them to suppose that 'there is "something" it is like to be us'- And it is a confusion to think, as Searle does, that for any conscious state, 'there is something that it qualitatively feels like to be in that state'.

10.3.3 The qualitative character of experience The attempt to capture the essential nature of consciousExperiences are specified by what , _. . t , . __ , r r , , , ness o r 0 1 what it is to be a conscious creature by means r they are of but not by reference _ _ . . , , . ,. ,.,..,., , ,i c i of the notion of there being something wbich it is like to to how theyjeel , , , , - , experience this or that or to be the creature one is has failed. Similarly, the notion that every conscious experience has a special 'feel' — that is, that there is a unique way it feels for a person to have any experience — has likewise proved misguided. Nevertheless, it may well be thought that less than justice is being done to those who seek to characterize experience in terms of its qualitative character. We have argued that it is licit to ask how it feels to have a certain experience or what it is like to have such-and-such an experience, and that these are actually questions concerning the subject's current attitudinal response to the experience he is undergoing. But, one may reply, this is not what was meant at all. So what was meant by the introduction of qualia? And is it coherent? It wili have been noted that the employment of the term Qtudia and the distinctive , , ,. . , „, . . r , , • of art quale is unstable. I he notion of a quale equivocr x 1 character of each experience . ^ ates between signifying whatever it is like for a person to have experience E and experience E itself. In view of our previous analysis, we must set aside the misbegotten phrases 'there is something which it is like' and 'there is something which it feels'. If we wish to get to the bottom of the concern with qualia, we must concentrate on the idea that every experience has a unique and distinctive character. Seeing red is different from seeing blue, and seeing a colour differs from hearing a sound or tasting a taste. So, too, feeling angry is different from feeling jealous, and both differ from feeling love or affection. W e noted that some writers attempt to extend the idea of qualia to thinking thoughts, holding (perfectly correctly) that thinking that 2 + 2 = 4 differs from thinking that 25 X 25 = 625, and conceiving of the difference (quite wrongly) as a matter of a difference in the qualitative character of the 'experience' of thinking. It is such differences that theorists have in mind when they misleadingly insist upon the unique qualitative character of conscious experience. Taken one way, this is both correct and innocuous. Of course, seeing red differs from seeing blue, and feeling love differs from feeling hatred. Of course, thinking that 2 + 2 = 4 differs from thinking that 25 x 25 = 625. Taken another way, it is confused. For the difference between seeing red and seeing blue does not lie in the way it feels or in what it is hke for a person to see the two colours. Yet these experiences do differ, and a normal human being who has such an experience knows perfectly well that the experience of

282

Consciousness and Contemporary

Neuroscience

seeing red differs from the experience of seeing blue, and is not likely to confuse the two. And whether or not thinking is correctly conceived to be an experience, someone to whom the thought that 2 + 2 - 4 occurs is hardly likely to confuse it with the thought that 25 X 25 = 625. So must not the difference between the experiences or between thinking the various thoughts reside in some quality of the experiences? And whatever this qualitative characteristic may be, it must be something that is apprehended by the subject of the experience, for it is this subjective apprehension that explains how the subject can differentiate the experiences he has. Or so it seems.

10.3.4 Thises and thuses We noted above that although many philosophers and . . , ., , . .. neuroscientists are taken with the notion of r qualia, and . . . accordingly insist that every experience has a umque qualitative character, none of them actually tells us, with respect to even one experience, what its specific character is. But it is striking that it is natural to try to refer to the specific quality of a given experience by means of an indexical expression, such as 'this' or 'that'. So we find David Chalmers asking 'Why do conscious experiences have their specific character?', and in particular, 'Why is seeing red like this, rather than like Í/ÍUÍ?'.34 And it seems evident that the 'like this' and the 'like that' are intended to be ways of referring to the specific qualities that experiences are alleged to have. _ , , So, human beings with normal visual capacities can see The logical constraints on such , , ,, , ,. , . . „ . , . , r red (green, blue, etc.) objects in their environment. Seeinaexical reference , , • ing a red object, we are told, has a particular subjective feel'. What is this 'subjective feel'? Well, seeing red is like this, seeing green is like this, seeing blue is like this - that is, this is the way I see red, this is how I see red. It if interesting, and striking, that Wittgenstein anticipated this confusion more than fifty yean ago. He wrote: Jndexical reference to the distinct, r , . we character of each experience

The content of experience. One would like to say 'I see red thus', 'I hear the note that you strike thus', 'I feel sorrow thus", or even 'This is what one feels when one is sad, this when one is glad', etc. One would like to people a world, analogous to the physical one, with these thuses and thises. But this makes sense only where there is a picture oí what is experienced, to which one can point as one makes these statements.35 His point is simple: it makes no sense to say 'I see red thus unless one can go on to sa} how I see red. W e labour under an illusion that when we see a red apple, we can, as il were, attend to our seeing, and say to ourselves 'This is how I see the red colour of the apple' and mean something intelligible, at least to ourselves, in saying this. But nothing meaningful is said, either to ourselves or to others, by saying 'this' or 'thus' unless there is a this oi a thus to which we can point — that is, unless there is a this or a thus in terms of which we 34 35

Chalmers, Conscious Mind, p. 5. Wittgenstein, Remarks on the Philosophy of Psychology, vol. 1, §896.

Conscious Experience, Mental States and Qualia

283

Cau

cash the sentence 'This is how I see red', or 'I see red thus', it makes perfectly good sense to say, 'I see the colour of the apple thus "^ I ' , pointing to a sample of red. Here the sample pointed at is what Wittgenstein, in the above passage, means by 'a picture' — that is, something that can represent, both for oneself and for others, how one sees the colour of the apple. But it is an illusion that one can, as it were, point inwardly (and for oneself alone) to the experience one is currendy enjoying, saying 'I see red thus', and thereby say anything meaningful - one might just as well say 'This is this' {see §3.9). It is 'as when travelling in a car and feeling in a hurry I instinctively press against something in front of me as though I could push the car from inside'. 36 If one thinks of perceptual experiences as thiscs and thuses, Incoherences that ensue when the . . . _, . , , t , . ,. , , it is tempting to go on to ask, as Chalmers does, Why logical constraints are disregarded . . , , . .r , .. a do conscious experiences have their specific character? - in particular, 'Why is seeing red like this, rather than like thai?. . . Why . . . do we experience the reddish sensation that we do, rather than some entirely different kind of sensation, like the sound of a trumpet?' 37 But now it should be obvious that the question 'Why is seeing red like seeing tin's ^ H [pointing to a red sample]?' is misguided. First, seeing red does not resemble seeing this "^ H colour; it is seeing this colour. Second, the only cogent answer to the confused question 'Why is seeing red like seeing this?' is that seeing this ^ H colour is seeing red, since this colour is what we call 'red'. Equally, the question 'Why, when one looks at red roses, does one not have the experience of seeing blue?' is a muddle. For the only possible answer (assuming normal vision and normal observation conditions) is trivial: namely, 'Because they are red, not blue.' What else would a normally sighted person expect to see when he looks at red roses in normal light? The concept of a normally sighted person is defined in part in terms of the ability to discriminate coloured objects. The visual system of normal human beings gives a person the capacity to discriminate between different colours, and normal human beings can distinguish between red and blue objects. We can investigate what features of our brains endow us with this capacity and what neural deficiencies deprive the colour-blind of it, and that is precisely what neuroscientists investigating colour vision do. There is no further question as to why when one looks at a red object in normal light one sees a red object. Even more misconceived is the question 'Why does one then have a reddish sensation rather than the sensation of the sound of a trumpet?' 38 The eye and the rest of the visual system evolved as a light-sensitive system endowing the animal with powers of visual discrimination. There is no such thing as seeing sounds with one's eyes. So there can be no puzzle as to why, when one looks at a red rose, one does not see the sound of a trumpet. Nor is it puzzling that, when one looks at a red rose, one does not, at the same time, hear the sound of a trumpet — given that no one blew a trumpet, and hence that there was no trumpet to hear.

L. Wittgenstein, The Blue and Brown Books (Blackwell, Oxford, 1958), p. 71. Chalmers, Conscious Mind, p. 5. Note that Chalmers is here using the term 'sensation7 to refer to perceptions rather than to sensations. Seeing red, strictly speaking, involves no sensations whatsoever.

Consciousness and Contemporary Neurosdence

284

Neuroscientists become similarly enmired in this morass _ . , c . , . , ,, c Crick, for example, characterizes the problem or quaha as 'how to explain the redness of red or the painfulness of pain', and observes that 'this is a very thorny issue', since the redness of the red that I perceive cannot be precisely communicated to another human being, at least in the normal course of events.39 But it is less than obvious what is meant by 'explaining the redness of red'. Are we puzzled why red things are red? One can ask why post-boxes in Britain are red — it is to make them salient. One can ask why blood is red - the haemoglobin, which contains iron, makes it red. These are no puzzles, and this, surely, is not what Crick means. But it is altogether unclear what he does mean, or, indeed, whether any intelligible question has been raised. Crick's puzzlement appears to be rooted in a pair of misconceived ideas. First, he appears to think that what I perceive cannot be communicated to another person. Secondly, he seems to think that colours are not properties of coloured things, but affections of the mind (so the redness I perceive is not a property of the poppies I am looking at, but rather the effect of those poppies on my sensibility — which cannot be compared with their effects on another person's sensibility). We shall revert to these misconceptions in the next section. So, what remains of the 'qualitative character of experiWhat the qualitative character of ,_, , v , ... . , „.,. , ... ence ? We must distinguish. With respect to any expelír an experience might be ° •/ ence we can ask what it was. The answer will specify the character of the experience — for example, whether it was feeling a twinge or a tickle, seeing a red rose or hearing the sound of music, feeling angry with A or jealous of B, playing cricket or going to the opera. We can also ask with respect to an experience what it was like to undergo it, and the answer, if there is one, will specify whether one found it enjoyable or unpleasant, interesting or boring, frightening or exciting, etc. None of this is mysterious, surprising or baffling. However, a false sense of mystery is engendered by the thought that the 'qualitative character of experiences' is incommunicable or only imperfectly communicable, that it is indescribable or only describable by reference to other experiences. It is to this ramifying confusion that we now turn. Crick's puzzlement

10.3.5 Of the communicabiHty and describability of qualia Whether the character of our experience is communic,, , .r , , .- ,. wr able and, if so, to what degree are matters ot dispute. We *T , , , , , , noted previously that Nagel holds that we do not possess a vocabulary adequate to describe the qualitative character of experience, as he conceives it. It is, he suggests, describable only in some respects, and such a description can be understood only by creatures like us. Edelman contends that 'qualia constitute the collection of personal or subjective experiences, feelings, and sensations that accompany awarene'"" They are phenomenal states . . . For example, the "redness" of a red object is a qualt The tndescribability and incom..... , municabthty theses

19

Crick, Astonishing Hypothesis, pp. 9f.

Conscious Experience, Mental States and Qualia

285

Qualia, h e holds, 'are e x p e r i e n c e d directly o n l y b y single individuals', a n d this gives rise to a methodological difficulty. For, ' W h a t is directly experienced as qualia b y o n e individual cannot be fully shared b y a n o t h e r individual as an observer.' But, h e suggests, the difficulty c^n be o v e r c o m e if w e 'assume that, just as in ourselves, qualia exist in o t h e r conscious h u m a n beings'. 4 0 G l y n n argues that W h e n we smell freshly ground coffee, or hear an oboe playing, or see the intense blue of a Mediterranean sky, or have a toothache, w e are having experiences that it is impossible to describe except by pointing to similar experiences on other occasions. W e can k n o w what it is like to have these experiences only by having them or having had them. There are all sorts of other things w e may be told about these experiences — what in the outside world is causing us to have them; what can be deduced about us or the outside world from the fact that w e are having them; what effect they are having or likely to have on our behaviour; what is going on in our brains while w e are having them - but n o n e of these things tells us about the subjective qualities of the experiences; what it is like for us to have them. 4 ' H u m a n beings w i t h n o r m a l c h r o m a t i c vision see red (green, blue, etc.) things in their environment. T h e y can, in d e c e n t lighting, distinguish those things that are red from those that are differently coloured. If they suffer from colour-blindness in its various forms, t h e n their discriminatory p o w e r s are deficient. B u t there are standard tests for colour-blindness. So w e can establish, w i t h a relatively h i g h degree of precision, w h e t h e r a n o t h e r person sees red, green or blue objects as normally sighted people d o . W h e n c e t h e n the illusion that so-called qualia are i n c o m m u n i c a b l e o r only defectively c o m m u n i c a b l e ? O n e source of the illusion is that I c a n n o t s h o w y o u m y _ , . , , - „ seeing. B u t w h a t is that supposed to m e a n ? Y o u can v ? \ . . certainly see that I see s o m e t h i n g . - Yes, it m i g h t be rephed, b u t you c a n n o t see what I see. B u t that is w r o n g , at any rate if w e b o t h have reasonably g o o d eyesight and are observing the same thing u n d e r optimal conditions, for then I can see exactly w h a t y o u see. — All right, o n e m i g h t respond, b u t at any rate, y o u cannot d o m y seeing for me] B u t w h a t does that mean? I can certainly see s o m e t h i n g , l o o k at something, instead of y o u , a n d r e p o r t back to y o u w h a t I see. — Yes, b u t for all that, you c a n n o t do m y seeing. B u t that is confused. F o r y o u c a n n o t do y o u r seeing either. Seeing is n o t s o m e t h i n g done. Y o u m a y see s o m e t h i n g , a n d I m a y or m a y n o t see w h a t you see. All that is true is (a) that it d o e s n o t follow from t h e fact that y o u see s u c h - a n d such that I t o o see s u c h - a n d - s u c h . H e n c e , (b) that the event of y o u r seeing s u c h - a n d - s u c h is a distinct event from the event o f m y seeing s u c h - a n d - s u c h , even if I d o see w h a t y o u see. A n d , (c) the fact that y o u see it in a certain w a y does n o t imply that I do, for w h a t you see clearly, I m a y see indistinctly, and w h a t strikes you as threatening m a y seem to me to b e altogether i n n o c u o u s . B u t there is n o t h i n g mysterious or i n c o m m u n i c a b l e about that. One source of the incommunkabil, . tty thesis

Edelman, Bright Air, Brilliant Fire, pp. 114f. Glynn, Anatomy of Thought, p. 392.

286

Consciousness and Contemporary Neuroscience

A second source of the confusion is one we have ahead . , 1 1 1 T .^ , Dennett, 'Consciousness', in R. L. Gregory (ed.). The Oxford Companion to the Mind (Oxford University Press, Oxford, 1989), p. 160.

r

298

Consciousness and Contemporary

Neuroscience

passive c o n c o m i t a n t of the possession of a sufficiendy elaborate control system, and dot not, in itself, actually " d o " anything'.6 If o n e starts from such misbegotten assumptions the • there is n o question b u t that the really deep p r o b l e m s lie h e r e . B u t these problems arise o r of conceptual confusions; and they are to b e resolved b y disentangling the knots that w have inadvertentiy tied in o u r understanding. . . , It is misleading to assert that the ... , , . world contains conscious states or

It is misleading to assert that it is a plain fact about tti ,, . • c • ¿ • . . . . _ A „ ,. , . r , conscious on the basis of inductive or analogical infería npkally direct, not inferred . . ° . " " ence, that 1 infer that they are conscious on the basis of ihcjr bare bodily movements, or that I assume that they are conscious, because they behave ,i jn pain, I should reply 'I saw him writhing in pain', and if asked how I know that he is enjoying himself, I should say 'I saw him joining in the fun with gusto'. But, first, the fact that I would justify my judgement thus does not show that I inferred it from the justifying evidence. For I do not reason; 'He is bleeding and groaning, there is a terrible grimace on [u> face; therefore he is, in all probability, in pain.' Nor is my judgement arrived at by means of an 'unconscious inference'. Rather, I recognize immediately and without any inference that the man is in agony. Secondly, the behaviour by reference to which one would The behaviour that justifies judge- . . - , . , , _,, , ,., , , , ¿-¡..ilustiry such judgements does not consist of bare bodily J ! merits about others mental states is J ' ° not 'bare bodily movement' movements , as the behaviounsts suggested, but of groans of pain, chortles of delight, tears of grief. "We describe such behaviour in terms of the rich vocabulary of psychology, and would be hard put to describe it in tenns of bare movements. (Try describing the difference between a smile of contempt, of amusement, of embarrassment, a cruel or kind smile, etc. in terms of the muscular movements. But none of us has any difficulty in recognizing such smiles in context.) There is nothing odd or unusual about this. We have the ability to recognize a pattern of behaviour in certain circumstances in psychological terms — that is, as a groan of pain, a laugh of joy, a sigh of relief. There is nothing more odd about this capacity than there is about our general capacity to recognize faces immediately and non-inferentiaUy. And, indeed, our ability to recognize behaviour as pain-behaviour or as behaviour of joy or affection is intimately related to our general capacity for facial recognition. For it involves, among other things, a sensitivity to, and recognitional capacity for, facial expression. Both powers have an evident evolutionary pay-off. , Thirdly, our judgements about others cannot rest on inr Judgements about others states of . . . . " . , . . . . , , . , Á. ducüve inference. For inductive inference presupposes nonr rr mind do not rest on inductive or , • , •f inductive identification of the relata antecedently found } analogical inference to be systematically correlated. It is because phenomenon

320

Consciousness and Contemporary Neuroscience

A has always been found to be correlated with phenomenon B that one is now wai in inferring from the occurrence of an instance of A the occurrence of an instanci , i But I obviously cannot identify another person's pain or joy, anxiety or good i independently of his behaviour, and find that such-and-such behaviour is well correlate | his pain or joy, etc. Rather, I am supposed to have noted that when 7 am in pam fete then I always behave thus-and-so, and then to reason that since others behave likewise circumstances of injury, they too are very likely in pain (etc.). But that presupposes (i) th I identify my pain, and then correlate it with my behaviour, and (ii) that I extrapolate fro i my own case. However, that is incoherent. For, first, I do not identify my pain (if I did, I miel In avowing and reporting my own . ., ,r . , , , . T ° ' • f ., .r • • mtsidentity it - and that makes no sense). I emnlov . v pain, neither identification nor en... . ' ' m t e n a o f l d e n t I t v i n smcerel teña of identity are involved Y a v o w i n S * * * have a pa . - I just have a pain and say so. But my ability to avc that I am in pain without any criterion of identity presupposes my mastery of the cor • fflon, public, concept of pain, including the behavioural (non-inductive) entena tr . justify ascribing pain to others. These are, as we have argued, two sides of one and t same conceptual coin. So not only do I not extrapolate from my own case when I asen pain to others, but rather, for there to be any 'my own case' (i.e. for me to be able employ the concept of pain in saying that I am in pain or in thinking that since I am pain Í had better take an aspirin), I must possess the concept of pain. And to possess t concept of pain, I must already know under what circumstances I am warranted in ascribing pain to others — that is, I must know the criteria for ascribing pain to others; namely, their pain-behaviour in appropriate circumstances. But these behavioural critena are not inductive evidence, but logically good evidence — which is partly constitutive of the concept of pain. So I have no need for inductive evidence in order to be able to asenbe pain to others. And if I did need inductive evidence, I could not have it, since I would not possess the concept of pain at all, and would not even be in a position to say that I am in pain whenever I am. Secondly, were I required to extrapolate from my own An analogical inference would rpre, ,, , . ., , , . ? . - . . case, what could warrant such an «responsible extrapoiasuppose identification in my own case r tion from one single case? So the putative inference cannot be inductive. Nor could it be an analogical inference? Do I infer that others are having a certain experience from their behaviour by analogy with my own case? This would, of course, make such alleged inferences much weaker than inductive ones. But, again, the supposition is not coherent. For just as with the putative inductive inference, the supposition of an analogical inference presupposes that in my own case I identify my own pain (or joy, depression, excitement, etc.) by using the concept of pain (joy, depression, excitement, etc.) independently of its nexus with the public criteria that warrant its ascription to others. But this means that I would have to have a criterion of identity to warrant its application to myself. But I have none. And the only way in which I could have one would be if I had a private mental sample of pain (joy, depression, excitement, etc.) against which to check the application of the concept to myself. But that, as we have argued (§3.9), is wholly incoherent, for there can be no such thing as a private mental sample which functions as a standard for the correct application of a word.

Puzzles about Consciousness

321

Is it then an assumption that others are conscious beings, n ;• not an assumption that other 1 , 1 • i 1 T i -, ^T It h " u . s that they have experiences as I do? N o , that too makes J «iji/fi are conscious r ' no sense, were it an assumption that they are conscious, rht'iJ it might be mistaken (as one might assume, and be mistaken, that a patient is conscious when he stirs after an anaesthetic and mutters something). But could anything count K showing that we are mistaken to take our fellow human beings, behaving as they normally do in the stream of life, to be conscious? When we see another person sorely injured and screaming in pain, do we assume that they are in pain? Is it a hypothesis? What more would we want, and what more could we have, to confirm this alleged assumption or In pothesis? More screams of agony? f he thought that we know that other human beings are conscious, and that they have Voracious experience' and are in 'conscious mental states', on the basis of an inference, inductive or analogical, or on the basis of an assumption, betrays the extent to which we ari" mesmerized by a misconceived picture, and manifests the manner in which we can [x-tome entangled in the network of concepts which, in our daily lives, we employ Linpi oblematically. We should remind ourselves of some platitudes. We react R,L„-t¿on ana response to others , .r . , c , , .r , , to the manliest experiences 01 others, to their manliest ¡m.c.les thought . \ . ' anger and gnei, their pain and joy, their amusement and excitement, instinctively — long before we are in a position to draw inferences from their Ivh.iviour, to make analogical inferences from our own case, or to make sophisticated nMumptions. A small child responds with fear to a display of parental anger, smiles with Jelibht in reaction to its mother's loving smiles and gende laughter, displays immediate unthinking anxiety in response to its mother's tears. These natural, instinctive reactions are the foundations upon which the gradual acquisition of psychological concepts rests. And mastery of the use of these concepts in the first-person case (which involves no evidence and no criteria of identity) is not detachable from the mastery of their use in the thirdperson case (which involves recognition of the behavioural criteria, which constitute logically good evidence for their ascription to others). The idea that the neonate, or its brain, has to 'construct' a model of its mother's mind (or brain) in order for it to 'predict' its mother's responses and behaviour is an absurdity which is part and parcel of a gross over-inteílectualization of human behaviour, of human empathetic and instinctive responses, and of the roots of language. What applies to ascription of consciousness and conscious It is neither an assumption nor a . , , , . ,. ,T , T , , . , , . , experience to other human beings applies no less to parallel hypothesis that other animals are . . . . , < , . ascriptions to the higher animals. It is not a reasonable r conscious beings ° assumption' or merely 'very probable' that monkeys, behaving as waking monkeys do, are conscious. After all, they are not asleep or unconscious. Do they not visibly see things? Are they not playful or angry? Do they not want things and try to get them? Indeed, are they not conscious of things in their vicinity (the food being brought to them at feeding time) and not conscious of other things (of another monkey creeping up behind them to snatch their banana)? This is no assumption, and there is nothing uncertain about it. Nor do we need the scientific community to swing decisively in favour of any such assumption in order to know perfectly well, as we do,

Consciousness and Contemporary

322

Neuroscience

that such animals are conscious beings. We can see the pleasure a dog takes in plai with a stick in the park with its master, as we can see its joy when it rushes to its returning master, wagging its tail and barking with excitement. So, too, we know perfectly well that the purring cat enjoys being stroked, and it is no less obvious that it is angry and threatening when it spits and raises its fur. , , , . , Our knowledge that the higher animals are conscious ua Knowledge that other animals are

.

.

, . ,

.

.

.

.

,

,

.,

beings is not based on an analogy with us, any more than J ian f , . our knowledge that human babies are conscious beings feel pain, perceive, are angry or pleased at this or that, is actually mere belief based on a shaky analogical argument from our own case. Human babies and animals alike satisfy the behavioural criteria for having sensations, for seeing and hearing, for feehng anger, fear and distress. It is visible that they are conscious and enjoy or suffer perceptual and affective experience. W e see that they do, and can justify our ascriptions of such experiences to them by reference to the behavioural criteria which they exemplify — not on the basis of an analogy with our own case. conscious is not analogical

12

Self- Consciousness

12.1 Self-Consciousness and the Self We have examined many conceptual features of con, , , „ , ,c , , sciousness, variously understood. But we have deterred thus . -, c ->c c i tar any investigation of the idea of serf-consciousness. is thorny topic has preoccupied psychologists, cognitive scientists and neuroscientists for ¡1 over a century. It has an even longer philosophical pedigree. The concern is with at we previously called the philosophical sense of 'self-consciousness', as distinct from the onion or garden variety pertaining to embarrassment and from the deliberative sense in ich we say of an artist or author that he is a highly self-conscious, reflective craftsman. I > related to the sense of'self-consciousness' in which we ascribe self-consciousness to an • -ospective person - that is, to a person who tends to reflect on his motives or reasons, • his likes and dislikes, on his character traits and his relationships with others. Such 1 erson is prone frequently to exercise a power which normal users of a developed language necessarily possess, but exercise relatively infrequently. To say that human beings are self-conscious creatures is not to suggest that we are all constitutionally introspective in the manner of a Proust. Rather, it is to say that we have the ability (mostly to a much lesser degree and with far less skill and refinement than Proust) so to reflect, as well as the more common ability, of which the former is a special case, to take into account, in our reasoning and behaviour, facts about ourselves, our experiences, past and present, and our character traits and dispositions. For we, unlike other animals, can be aware of the fact that we are feeling cheerful or depressed, that we are ignorant or well informed about some subject, that we have certain beliefs or doubts, that we have certain character traits and dispositions, that we have done and undergone various things in the past. Such facts about ourselves, of which we are cognizant, may weigh with us in our deliberations and occupy us in our reflections; they may be our reasons for acting, feeling or thinking thus-and-so here and now. They may be premisses in our reasoning and our justifications for our actions and reactions. This is partly constitutive of our being self-conscious creatures. Deep conceptual confusions bedevil the debate about the oelj-conscwuness and 'the self ._ _ . ., . , r nature of self-consciousness, r o r philosophers and nonphilosophers alike are prone to construe self-consciousness as consciousness of something '-consciousness in the phtlo, deal sense

324

Consciousness and Contemporary Neurosaence

they call a 'self, an T or an 'Ego'. This entity is supposedly possessed by each person wh is said to have a self — I have my self, and you have your self My self is thought to \something 'in me', of which I can become aware or of which I am allegedly constant1 aware. But when philosophers and psychologists have tried to pin down this curioi entity, they have found it singularly elusive. And recent efforts by neuroscientists to she light on the matter have been equally unsuccessful - or so we shall argue. This is llt surprising, for, as we shall show, the 'self1 or the T (thus conceived) is a fiction generate by conceptual confusions. There is indeed such a thing as self-consciousness in the plu] t sophical sense of the term, but it is not consciousness of a 'self. It is a distinctively huma capacity for reflexive thought and knowledge, wholly dependent upon possession < a language. Before examining recent scientific reflections on this vexed subject, a brief survey oft] roots of the philosophical tradition is appropriate. For many of the confusions of neun scientists, cognitive scientists and psychologists recapitulate the errors of past philosophy

12.2 Historical Stage Setting: Descartes, Locke, Hume and James From Descartes onwards, philosophers have been niucn The Cartesian ego as an immate, ., . , ., , r , . , , * concerned with trying to elucidate the nature or the rial substance ,. ^ , , , subject of experience. Descartes argued correctly tint ,i thought cannot exist without a thinking thing - a substance (a persistent entity that is ifte bearer of attributes) in which to inhere. Awareness of his own thoughts, he claimed, proved his existence as a substance, which is named or signified by the word T . But what this T is must be investigated, since 'I must be on my guard against carelessly taking something else to be this " I " and so making a mistake in the very item of knowledge tliat I maintain is the most certain and evident of all'.' For while he could, he thought, doubt whether the material world exists, and hence too whether his body exists, 'I cannot doubt whether I exist', for in order for doubting to occur, there must be some thing that is doing the doubting. And he concluded that 'this " I " - that is the soul by which I am what I am - is entirely distinct from the body'. 2 It is, he held, an immaterial thinking substance, intimately conjoined with, but distinct from, the body. We should note immediately that the phrase 'this " 1 " ' is The ungrammaticality of 'the "T" . , £ j . . . ,, ~ . or German. But, first, talking to oneself inwardly need tion is neither necessary nor sufficient .. ° . ' , ,•,• involve no thinking. Reciting the multiplication tables to oneself in one's imagination, or going over one's speech to make sure that one knows it by heart, or 'counting sheep' in order to stop oneself from thinking and to enable one to fall asleep, are not forms of thinking. Secondly, one can think without talking to oneself in the imagination at all. One can come to the conclusion that p on the basis of evidence e, or see that what follows from a and h is that c, without saying anything to oneself; all that is necessary is that one thenceforth be willing, other things being equal, to assert that p on the basis of the evidence e, or to assert that c for the reason that a and b, or be willing to act for the reason that p and be able to cite the fact that p as one's reason for doing what one does. , r < , ,, , , We may be inclined to think otherwise because we are On the use of to be able to think . , , . , . . . , , over-impressed, or wrongly impressed, by the fact that we say such things as '1 can speak in German, but I cannot think in German'. In so saying, one signifies that before one can say something in German, one must, by and large, first decide what one wants to say (and be able to say it in English), and then struggle to find the right German words. It does not follow that it makes sense to say of a native English speaker that he thinks in Enghsh, unless that just means that when he talks to himself in his imagination, what he thus says to himself is in English. Of course, one also says of an Englishman that he speaks German so well that he even fíiínfes in German. But that, if it does not mean that he speaks to himself in his imagination in German, just means that he does not first think of what he wants to say and then pause to try to think of the German words. . , , , We are doubtless also deceived by the analogy between Hunting for the right word ^ ,• , , , - r i - i ^, . , ,, , an Enghsh speaker hunting for the right German word in - a misleading analogy , f _° order to say such-and-such and an Enghsh speaker hunting for the right word in English to express his thought. But the analogy is deceptive. For in the first case, he can say, in English, what he thinks; but in the second, he cannot. This is not because he has thought in images and has not found the right translation, so that, so to speak, he knows what he thinks and is now looking for the correct words to express his (subjectively) wholly perspicuous thought. 'The word is on the tip of my tongue' means much the same as 'The right word escapes me for the moment, but it will come to me shortly, I hope'. Similarly, 'I know exactly what I want to say, but I can't think of the words' is either nonsense, or it means no more than 'Give me another moment for the thought to crystallize and then I'll tell you what I think'. So, too, when someone else finds the right words — the words I was looking for to express my thought — the rightness of

344

Consciousness and Contemporary

Neuroscience

their w o r d s does n o t consist in their being a correct translation of m y t h o u g h t or in their correctly matching m y wordiess t h o u g h t . R a t h e r , it consists in their m a t c h i n g the p h e n o m e n o n w h i c h is the subject of the t h o u g h t i n a way w h i c h strikes m e as appropriate, and w h i c h I was trying to capture myself. W h a t , t h e n , o f the observations o f Einstein, Galton , _ „ r . . . , , , ' i r , Hadamard and Penrose? barfrominsisting that they think , , 7 in language, they all insist that they d o n o t . Einstein insisted that the elements o f his t h o u g h t w e r e 'certain signs and m o r e or less clear images 1 ; Penrose asserts that h e thinks 'visually' i n terms of 'specially designed diagrams' w h i c h are a s h o r t h a n d for algebraic expressions. G a l t o n stated that h e h a d to translate his thoughts into language. D o e s this n o t s h o w that o n e thinks in images, a n d t h e n translates one's t h o u g h t s i n t o language? It does n o t . W h a t it shows is that the description o f w h a t goes o n i n one's m i n d w h i l e o n e is t h i n k i n g is typically a description neither of one's t h i n k i n g n o r of w h a t o n e is thinking. W h a t o n e is thinking is n o t w h a t is present t o the imagination while thinking, save i n cases i n w h i c h w h a t one says t o oneself is w h a t o n e is thinking. A n d a description of w h a t passes t h r o u g h one's m i n d w h i l e t h i n k i n g is n o t a description of one's thinking. O f course, s o m e people m a y i n v o k e mental images as Mental images may be heuristic , . . . . . . . ,. ,. . , , , &, ' , , heuristic devices t o aid t h e m i n thinking, mst as others devices, but they are not thoughts •,,,•,• 1 1 r ,. m a y scribble diagrams or fragmentary symbols o n a piece r i( r or expressions oj thoughts ' ° ° ' ' of paper. N e i t h e r the o c c u r r e n c e of the images n o r the w r i t i n g of the scribbles are the thinking. N o r are the images or scribbles expressions of the t h o u g h t . T h e mental image o n e conjures u p may be a mental image o f w h a t one is t h i n k i n g about, b u t it c a n n o t b e what o n e is t h i n k i n g {viz. that p). M e n t a l images m a y b e aids t o t h o u g h t , often essential aids t o t h o u g h t — b u t a description of the succession of these images, including images o f diagrams a n d / o r algebraic symbols, w o u l d n o t b e a description either of the person's thinking (which m a y b e described as swift, insightful, and impressive, or slow, clumsy and ineffective) or a statement of what h e was thinking. And there is n o such thing as translating these images into language. Gallon's insistence that h e did n o t think in words is There is no mystery about not ,. T , , .. , . . , .... ,, ,, ; . . perfectly in order a n d unsurprising. W h a t it means is that thinking in words , but it is con' ,- ,,, , ,, , ,• , ,¿ , , , ,, w h e n t h i n k i n g his w a y t h r o u g h a problem, h e did not r & jused to suppose that one s thought / is r talk t o needs 'translating' into words himself m his imagination. W h e n he reached a conclusion, h e k n e w that h e h a d solved his p r o b l e m and k n e w that h e could spell o u t the solution, even t h o u g h h e h a d n o t d o n e so. T h a t is n o m o r e mysterious than k n o w i n g w h a t o n e is g o i n g t o say before one says it, w h i c h o n e normally does - it w o u l d b e truly mysterious if o n e n e v e r k n e w w h a t o n e was going t o say before o n e said it! B u t k n o w i n g w h a t o n e is g o i n g t o say does n o t mean saying it t o oneself before o n e says it out loud. N o r does it imply t h i n k i n g w h a t one is going t o say in a non-linguistic m e d i u m of t h o u g h t . W h e r e G a l t o n is altogether misleading is in his assertion that h e had to translate his 'wordless t h o u g h t s ' i n t o language. For k n o w i n g a solution t o a problem, and saying w h a t t h e solution is, is n o t translating anything. It is actualizing an ability - namely, the ability to give the right answer to the problem. On so-called non-linguistic thought _. . _ , * , . j - Einstein, Galton, Hadamard and p

T I 1

Self- Consciousness

345

Similarly, Penrose is misleading n o t in asserting that h e , , . . ,. , , . . , . . , , , does n o t t h i n k in w o r d s , b u t m claiming that there are n o t the w o r d s available to express the concepts that are required'. F o r mathematical symbols express concepts, have a r u l e - g o v e r n e d use, n o less than w o r d s . T h e w o r d s 'cat', 'chat', a n d 'Katze' are symbols in three different languages, all of w h i c h express o n e and the same concept; the arithmetical symbols ' + ' , ' = ' , a n d Y are universally accepted symbols in the 'language of arithmetic' expressing the concepts of the addition function, of equality and of the square r o o t of minus 1. T h e r e are w o r d s to express mathematical concepts. It w o u l d b e extremely laborious, h o w e v e r , to use those words rather than t h e mathematical symbols, a n d quite impossible to m a k e c o m p l e x calculations w i t h o u t these symbols. N o d o u b t , w h e n Penrose finds a mathematical solution t o a p r o b l e m , w o r d s d o n o t piay any significant role in w h a t goes o n i n his m i n d . Mathematical symbols and imagined diagrams m a y play a heuristic role. B u t w h e n h e has solved a mathematical p r o b l e m , h e can write d o w n the solution in mathematical symbols, and these express concepts, mathematical concepts, n o less than d o words of natural language. So, is language necessary for t h o u g h t ? T h e question is Reformulating the question . , , . , , , J t o o c r u d e , a n d needs t o b e b r o k e n d o w n i n t o a s e q u e n c e Confusions about words and J

of questions. First, can creatures w h o have n o t mastered a language , . , . ,„, , , , , , . . n ,• think? W e have already answered this question. R u d i mentary t h o u g h t can b e ascribed to n o n - h u m a n animals, b u t only to the extent that w h a t they are said to think can b e manifest in their behavioural repertoire. . , Secondly, in order to think, does o n e have to think in The question of whether one has to , - „r , , , ,. . . . , , ,. , , . . , ,. language? W e have suggested that this question is m i s think m language is misleading * ^ °° , . . . leading. O n e m a y talk to oneself i n o n e s imagination while o n e is thinking, or o n e m a y not. O n e m a y silently rehearse one's t h o u g h t s in an internal m o n o l o g u e , or o n e m a y not. B u t talking to oneself in the imagination is n o t the same as t h i n k i n g a n d n e e d involve n o t h o u g h t . A n d if o n e does n o t say w h a t o n e thinks to oneself in the imagination w h e n thinking, it does n o t follow that o n e must be t h i n k i n g in some non-linguistic m e d i u m , such as images. T h e v e r y phrase ' t h i n k i n g in language' leads us astray, for it is apt to b e construed o n the m o d e l of 'speaking in English' or 'speaking in G e r m a n ' . O n e n e e d n o t think in a n y t h i n g (in this sense), for o n e n e e d n o t talk to oneself w h e n o n e is thinking, a n d even if o n e does talk to oneself while thinking, what o n e says to oneself m a y n o t b e w h a t o n e is t h i n k i n g at all (as Penrose rightly points out). T h i r d l y , can o n e think in images'? T h i s question t o o is The question of whether one can . , ,. ,, , , . , . , , ., ,. , ,. . , , . , ,. misleading. A l t h o u g h images m a y cross o n e s m i n d w h i l e think in images is misleading . , . . . 111 o n e is thinking, a n d although o n e m a y use images heuristically, neither the images n o r their descriptions are expressions of the t h o u g h t . Further, the process of images passing t h r o u g h the m i n d is n o m o r e the process of thinking than the succession of heuristic scribbles o n a piece of paper are the process of thinking. In so far as t h i n k i n g (reasoning) can be said to b e a process, it is a process of reasoningyrom s u c h - a n d - s u c h premisses to a certain conclusion. T h e train of t h o u g h t is laid out in the explicit statement of the inferences, n o t in the description of the images, Can non-human animals think?

346

Consciousness and Contemporary

Neuroscience

symbols or diagrams that crossed one's mind while one was thinking or that one used as heuristic devices to help one think one's way through to the solution. Having an image of X before one's mind, is not thinking in images in the sense in which one speaks in English. The words one utters when one speaks are the expression of one's thought. The images one conjures up while thinking are not the expression of one's thought but an aid to thought or an accompaniment of thinking. Finally, must one have mastered a language in order to The limits of possible thought are . ,, ... ,. , , , ,. ... A . .. . / , ... be able to think anything beyond the rudimentary thinktne limits of the possible expression n , , T > r,, ,. ing of an animal? That is indeed what we are suggesting. For, to repeat, the limits of possible thought are the limits of the possible expression of thought. A thinker can think only what he can (but need not) express (or what he would have been able to express, had he not lost the power of speech which he previously had). An Einstein need not talk to himself while he is thinking. Nor need anyone else. But he can think only what he can express (or recognize as being expressed) — in words, symbols, diagrams or formulae.

12.6 Self-Consciousness A non-human animal can feel sensations, perceive things, feel emotions, want things, and act in pursuit of what it wants. It can know a variety of things, and, in a rudimentary sense, it may think or believe various things. It is, as we are, either conscious or unconscious; it enjoys perceptual, affective and volitional (conscious) experiences; and it can become and be conscious of various things inasmuch as its attention may be caught and held by items it perceives. Nevertheless, it is not a self-conscious creature. It can perceive, but it cannot think about or reflect on the fact that it is perceiving whatever it perceives. It can be angry, frightened, jealous, affectionate, excited; it has likes and dislikes, and can take pleasure in a range of activities. But it cannot realize, and so become conscious of the fact, that it is frightened or jealous, or that it is excited or taking pleasure in its activities. Hence, too, the fact that it is feeling anxious or terrified cannot occupy it and weigh with it in deliberation and intentional action, even though its actions are affected by what it is feeling. Even more obviously, the fact that it has, for example, an irascible or affectionate disposition cannot be something of which it may become and then be conscious. Nor can it reflect on its past experience, even though its past experience may affect its current behaviour and reactions. That an animal is not a self-conscious being is not to be Self-consciousness is not conscious, . ,, , ., , . , , < ,r, c ,r . explained by reference to the idea that it lacks a selr , or r ness of a sen; it presupposes con' „ . , ,.,, that it has a self b u t is n o t conscious o f its self . For, ceptual skills as we have argued, there is no such thing as 'a self conceived as the inner owner of experience;'and the fact that we are self-conscious beings does not consist in our having and being conscious of'a self' either. What an animal lacks, and what we have, is mastery of a language. So, we agree, roughly, with Edelman, who insists that what he calls 'higher-order consciousness', by contrast with 'primary consciousness', Consciousness contrasted with self-consciousness

Self- Consciousness

347

requires a language. But this is not because a language is necessary inasmuch as 'A conceptual model of selfhood must be built, as well as a model of the past'.28 Nor because 'animals, having only primary consciousness also have qualia, [but] cannot report them explicitly either to a human observer or to themselves, for they lack conceptual selves'.29 But rather, because in order for the fact that one is enjoying an experience of a certain kind to strike one, in order for one to be able to think about it, and in order for it to weigh with one in one's deliberations, one must possess the concept of the relevant experience. To realize, and so become conscious of the fact, that one is jealous, angry or afraid, and for that fact to occupy one's thoughts and to constitute a reason for one to do or think something or other, one must possess the concepts of jealousy, anger or fear. Similarly, for one to be conscious of one's ignorance or erudition, of one's vanity or pride, one must possess the concepts of these characteristics. And for one's past to weigh with one in one's reflections, one must be able to think about it — and that presupposes mastery of language. Human beings, but not other animals, have an autobiography. And human beings, but not other animals, may have a sense of history, may be — and typically are — conscious of themselves as historical beings, as belonging to a society or social group with a certain history. Self-consciousness, we have argued, does not involve Self-consciousness requires mastery . . . / , consciousness of a self. But it does involve masteryJ of the of personal pronouns . use of personal pronouns in general, and of the use of the first-person pronoun in particular. To master the use of T does not require that one notice an inner object which one had not noticed hitherto, which the pronoun T names. For there is no inner object called 'the " 1 " ' or 'the Ego', and the first-person pronoun is not a name. To learn the use of the first-person pronoun, one does not have to learn to identify an object of any kind, either inner or outer. The characteristic use of T admits of no misidentification or misrecognition. 30 But that is not because it always involves an unerring identification and recognition of oneself (let alone of one's self). It is rather because no identification or recognition is involved at all. But mastery of the use of the first-person pronoun by a child in the course of normal human development does involve, and goes hand in hand with, mastery of the use of other personal pronouns and personreferring expressions. As we argued above, the transition from inchoate cries of pain to 'Hurts!', 'It hurts!', and then to 'I have a pain' is bound up with understanding the question 'Does it hurt?', and hence too with learning the use of the interrogative 'Does Mummy have a pain?', and so with mastering the use of the declarative sentence 'She has a pain'. For in learning to give verbal expression to his own pain, the child also learns to describe others as being in pain — for the first-person and third-person pain-predications are two sides of one and the same linguistic coin. One cannot be said to possess the 28

Edelman, Bright Air, Brilliant Fire, p. 131. Ibid., p. 135. There is a possibility of misidentification in rather special cases: e.g. when looking at an old photograph, one may point at a figure, and say 'I was quite a pretty baby', and one may be wrong, inasmuch as the photo is not of oneself. This is a case of misidentification - but it does not affect the 2¡}

above argument.

348

Consciousness and Contemporary Neuroscience

concept of pain unless one has mastered both its criterioniess use in the first-person case and its criteriaily based use in third-person ascriptions of pain (see §3.9). Similarly, one cannot be said to have mastered the use of 'I . . .', unless one has grasped that saying 'I am . . .' is a ground for others to say of me 'You are . . .' or 'He is . . .'. The first-person pronoun is one. piece in a complex game in which the other personal pronouns and person-referring expressions are other essential pieces. Like the king in chess, it is the pivotal piece for each player; but without the other pieces, one cannot play the game. The pattern that we are invoking in explaining mastery I JF P P 0 £ t k e firS(;_person pronoun can be applied similarly to first-person perceptual utterances, such as 'I see . . .' or 'I hear . . .', etc. The learning child must first have mastered the use of appropriate descriptive terms to describe what he perceives - for example, 'Teddy is on the floor', 'Rover is barking', and so forth. To learn to prefix an 'I see . . .' or an 'I hear . . .' to such descriptions is again not to learn to identify some other object named by T . Rather, the child must have learnt that seeing can be cited as one way of finding out that things are thus-and-so (e.g. that Teddy is on the floor, that the bail is red, that the sun is round), as hearing can be cited as the way to find out other things. And this he will obviously learn as his parents ask him 'Can you see where Teddy is?', 'Can you hear Rover barking?', and so forth. The conditions that warrant prefixing an 'I see . . .' or an 'I hear . . .', the child will learn, are no different from the conditions for asserting what one sees or hears by using ones eyes or ears respectively. The prefix indicates to another how one is or came to be in the position to assert what follows: that is, how one knows (or thinks one knows) that things are as one describes them as being. But in gradually mastering the use of the firstperson pronoun in such contexts, the child must, and unavoidably will be taught to, master the use of the second- and third-person pronouns in these contexts as well. He will learn to ask 'Can you see?' and 'Did you hear?, and to report 'Mummy didn't see it' and 'Daddy didn't hear' when he sees that his mother failed to notice something or that his father was not listening. The possibility of the various forms of self-consciousness is therefore acquired together with, and not antecedently to, the mastery of the linguistic apparatus for describing the experiences of others. Hence the idea that the child first learns to ascribe experiences to himself, and only then to ascribe experience to others, on the basis of analogy with his own case or on the basis of his construction of a theory about 'other minds', is utterly misconceived. It is misleading to characterize such processes of masterSelf-consdousness is bound up with . , , . . , , . , J L . ,. . , . , 1 ) m e the linguistic techniques that are presupposed by selt-l f soaahzation, hut not with model ° ° ,,.,,. , , , c m .> , .,,. consciousness as bunding a conceptual model ot selinooa or 'constructing a socially based selfhood', as Edelman suggests. But it is obvious that the possibility of self-consciousness is bound up with rich and complex forms of infantile socialization. The young child responds to parental emotion instinctively; he reacts, without thought or inference, to his parents' loving caresses, to their anger, to their approval and disapproval, to their smiles and their tears. The idea that in order to do that, the child (or his brain) must 'construct a model' of his parents' minds, which model will enable him to predict their behaviour, is surely as preposterous as the idea that the fledgling, kitten or puppy constructs a model of its parents' minds.

Self-Consciousness

349

Model building is an activity that presupposes sophisticated conceptual abilities, and we cannot invoke it to explain their acquisition. Furthermore, it is literally unintelligible to ascribe it to the brain, if 'model building' has its normal meaning in the context of such ascriptions. Self-consciousness is not a 'construct' of any kind, but an ability. It is not evident what 'selfhood' is supposed to be. But if what one has in mind here is the emergence of the child's conception of himself, and abihty to think of himself, as a child, as the son of such-and-such parents, as (to put it briefly) the subject of such-and-such predicates, then, to be sure, the child's 'selfhood', or, more perspicuously, the child's sense of himself as a child of whom such-and-such is true, emerges with his socialization and his growing linguistic skills. It is inseparable from his mastery of the first-person and otherpersonal pronouns, of the criterionless use of psychological predicates in the first-person case and of their criterial use in the third-person case, and from the very large range of other linguistic skills he must already have mastered before he has reached that stage of linguistic maturation. . . r ,r The capacities of selt-consaousness

A creature that has thus mastered a language is a self. , ,, . . , 1 conscious creature. It not only has conscious experience (and is conscious or unconscious, and may be conscious of this or that), but also has the ability to give articulate expression to its thoughts and experiences. It can not only perceive, it can say that it perceives; not only feel, but also say what it feels. It can think, and not say what it thinks, precisely because it can say what it thinks. Furthermore, it can not only think and give articulate expression to its thoughts and experiences; it can think of itself as having those thoughts and experiences. It is precisely because it can give articulate verbal expression to its thoughts and experiences as its thoughts and experiences, that it can also reflect on its thoughts and on the fact that it is thinking such thoughts. It can think about its experiences and about the fact that it is having such-and-such experiences - even though it may not give any overt articulate expression to its reflexive thoughts and ruminations. In short, we not only perceive, feel, want and think whatever we do; but we can say that we do. And inasmuch as we can say what we can thus say, we can also think about our condition and about ourselves as being in that condition without saying anything. This ability is constitutive of self-consciousness, and ít is a prerogative of language-using creatures. lt w o u l d b e a i m s t a k e t o s u The scope of the powers of P P o s e t h a t t h e P o w e r s o f self ~ self-comciousness consciousness are restricted, in respect of their objects, to one's current thoughts, mental states and experiences. A self-conscious creature such as a human being can be conscious of his own motives, and that he is acting for such-and-such motives may occupy his thoughts and weigh with him. Often one becomes conscious of one's motive in action by realizing that what moves one are such-and-such reasons, and that these reasons fall into the pattern of a certain motive - for example, charity, love, jealousy, vindictiveness, ambition, revenge, fear. As we have seen (§7.2.2), the pattern, to which we commonly give a motive name, consists of a backward-looking reason for action - for example, some past or present fact that is undesirable in a certain way — and a forward-looking reason for action specifying a future state of affairs that removes or compensates in some way for the undesirable state of affairs. To act out of love for Mother is (roughly) to conceive of Mother's condition as lacking in

350

Consciousness and Contemporary Neuroscience

some way, to view a future state of afíairs that can be brought about by a certain action as beneficial to Mother, and to do that action with the intention of benefiting Mother for her sake. To act out of revenge is (roughly) to believe that, for example, Jack harmed an interest one had, to view a certain action as harming Jack, and to perform tiiat action with intention of harming Jack because he harmed an interest one had. Clearly, one can act out of a certain motive without having a concept of that motive. But one cannot realize, and so become conscious of, one's motive in acting unless one has a concept of the relevant motive. Having realized one's motive, one may be pleasantly or shamefully conscious of the fact that one is acting out of such a motive. So, too, one may be unclear as regards one's true motive in action, and one's lack of clarity may weigh on one and occupy one's thoughts — and this too is something possible only for a self-conscious being. Self-knowledge involves knowledge of one's own character traits, dispositions and skills (or lack of skills). What these are, one may come to know in various ways: for example, by being told by others, or by realizing that one's past behaviour amounts to being timid, idle, vindictive, fastidious or boorish. One does not come to know one's traits and dispositions, however, by being perceptually or 'introspectively' conscious of them. For one's character traits, dispositions and skills are not objects that one can perceive, let alone 'introspect' (misconstrued as apprehension by 'inner sense'), and become conscious of. But knowing that one has them, one may be conscious of them - that is, conscious of the fact that one is thus-and-so. And if one is conscious of them, they occupy one's mind and weigh with one in one's deliberations. This too is only possible for a language-using, self-conscious creature. For in order to be conscious of my ignorance or erudition, my prowess or incompetence, my vanity or ambition, my forgetfulness or weakness of will, I must possess concepts of the relevant characteristics, the self-ascription of which is involved in my being conscious of myself as possessing them. Self-conscious beings such as ourselves know, and maybe , , . , , , r conscious of, not only their current thoughts and expenJ r . enees, motives and traits, etc., but also their past history. As remarked above, human beings have, and animals do not have, an autobiography (and, to a greater or lesser degree, an awareness of history). Hence what a person knows or believes about his past (and the past of his society, social group or family) is something of which he will be dispositionaily conscious, and of which he may become occurrently conscious (see §9.3) when such facts or putative facts occupy him in thought, deliberation or fantasy. Such facts or beliefs make up a large part of his sense of himself (but not of his self). For they are partly constitutive of a person's sense of identity as a person of suchand-such a kind, with such-and-such allegiances, commitments, obligations and rights. „ ,r . . , . . It should now be evident how misconceived is the idea ¿>etf-consaousness is not explained , ,, . , , ,~ , ,r , . that selt-consciousness is, or must involve, a neural seltc by reference to selj-scanmng devices in the brain scanning or self-monitoring device in the brain. Whether the neural networks of the brain contain anything remotely resembling a self-scanning device that .might be built into a computer, we do not know, and nor does anyone else. But even if they do, why should we suppose that this would clarify the nature of self-consciousness? It is clear enough that the suggestion that it Having an autobiography and ... . * * J sense of identity

Self- Consciousness

351

ivould is predicated upon a conception of self-consciousness as a form of introspective scanning of the contents of the mind. If one thinks thus, then one might also think that to make self-consciousness possible, there must be a correlative self-scanning device in the brain.31 But this is doubly misconceived. First, as we have argued, self-consciousness is not a matter of scanning anything in 'inner sense' (i.e. by introspection, wrongly conceived (see §3.6)). Secondly, even if {per itnpossibile) it were, it is obscure how a neural self-scanning device could contribute anything to the possibility of self-consciousness. A neural self-scanning structure in the brain would no more be conscious of brain events than a computer program, which incorporates as a subroutine a 'description' of another program, is conscious of that other program. A self-referential component in a program cannot make a computer conscious of anything, let alone of itself. Aiming a video camera at a mirror, as Penrose observes, does not make it conscious of itself.32 So even if the brain does contain some analogue of a selfscanning device, the results of its 'self-scanning' could not be something it can communicate to the person whose brain it is. And if it could, the person in question would not be able to understand it, unless he were a 'super-neuroscientist'. For the most that a neural selfscanning device could reveal would be complex neural structures. It may be that a person would not be able to do some of the things he can do, and would not be capable of the various forms of self-consciousness, unless complex, selfscanning neural networks were operating. But if that is so, it is not because self-conscious reflexive thought involves scanning anything. There is no special reason to suppose that a person's ability to think about himself, about his current perceptual experiences or his past ones, about his current emotions or motives for action or his past ones, or about his character traits and dispositions, is dependent upon a neural seif-scanning device that scans something that is, or is systematically related to, what the person is conscious of. ..,.,, ., , , Linguistic skills provide the key , ,. , , c to understanding the nature of ./• .

It is, to be sure, remarkable that there are any self. „ . , , . conscious creatures at all. Certainly, we would not be ' the kinds of creatures we are, were we not self-conscious beings. To be fully human is to be self-conscious — that is, to have the ability to engage in the manifold forms of reflexive thought and reflection. It is therefore also the ability to take facts concerning one's own beliefs or knowledge, experiences and dispositions, emotions and motives, past history and social relations, as reasons for thinking thus-and-so, grounds for feeling such-and-such (including feelings of moral fulfilment, or of remorse or guilt), and as reasons for action. But the key to the understanding of self-consciousness, its nature and possibility, lies not in neural selfscanning mechanisms in the brain, but in the normal human mastery of language. And the key to the understanding of its neural foundations lies in an understanding of the neural conditions for the possession and exercise of linguistic abilities and of the neural conditions for the possibility of the forms of thinking that are a consequence of the possibility of speaking. These may, or may not, involve neural self-scanning devices. This is strikingly evident in Weiskranz's explanations of blind-sight; see below §14.3.1. Penrose, Emperor's New Mind, p. 530.

Part IV On Method

13

Reductionism

13.1 Ontological and Explanatory Reductionism M a n y brain-neuroscientists have an implicit belief i n , . _ . , , , , reductionism. h e w try to articulate w h a t exactly thev , , * , , , r m e a n by this t e r m 01 art. A m o n g those w h o d o , the m o s t lucid statement of the c o m m o n c o n c e p t i o n of r e d u c t i o n i s m w i t h respect to cognitive neuroscience that w e have found is given b y Francis C r i c k in his b o o k The Astonishing Hypothesis. Indeed, o n e reductionist thesis that C r i c k defends is also w h a t h e holds to b e the e p o n y m o u s 'astonishing hypothesis': namely, that ' " Y o u " , y o u r joys a n d y o u r sorrows, your m e m o r i e s and y o u r ambitions, y o u r sense of personal identity a n d free will, are in fact n o m o r e than t h e b e h a v i o u r of a vast assembly of nerve cells a n d their associated molecules.' 1 It does n o t , C r i c k avers, ' c o m e easily to believe that I a m the detailed behaviour of a set of n e r v e cells', b u t in fact it is so. This c o n c e p t i o n appears to b e a f o r m of ontological reductionism, i n a s m u c h as it holds that o n e k i n d of entity is, despite appearances to the contrary, actually n o m o r e than a structure of o t h e r kinds of entity. Side b y side w i t h t h e ontological reductionism, C r i c k also defends a form of explanatory reductionism: ' T h e scientific belief is that o u r m i n d s — t h e b e h a v i o u r of o u r brains — can be explained b y the interactions of n e r v e cells (and o t h e r cells) and the molecules associated w i t h t h e m . ' T h e reductionist approach, C r i c k explains, is that Ontological and explanatory , . , ,. . . , , _ •i reductionism distinguished: Crick

a complex system can be explained by the behaviour of its parts and their with each other. For a system with many levels of activity, this process may repeated more than once — that is, the behaviour of a particular part may explained by the properties of lis parts and their interactions. For example, to

interactions have to be have to be understand

1 F. Crick, The Astonishing Hypothesis (Touchstone, London, 1995), p. 3. It is surprising that Crick should find his materialist hypothesis 'astonishing', since it was already propounded in Epicurean atomist form in the first century BC by Lucretius in his great poem De Rerum Natura. In somewhat different forms, it was defended by Gassendi and Hobbes in the seventeenth century, and by La Mettrie, Diderot and d'Holbach in the eighteenth. So, if it is astonishing, that is certainly not because of its novelty.

356

On

Method

the brain w e may need to know the many interactions of nerve cells with each other; in addition, the behaviour of each nerve cell may need explanation in terms of the ions and molecules of which it is composed. 2 R e d u c t i o n i s m , Crick holds, is 'the main theoretical m e t h o d that has driven the development of physics, chemistry a n d molecular biology. It is largely responsible for the spectacular developments of m o d e r n science. It is the only w a y to p r o c e e d until and unless w e are confronted w i t h strong experimental evidence that d e m a n d s w e modify o u r attitude.' 3 „, , The consequences that are , . , r,, , sometimes derived: Blakemore

C o l i n Blakemore p r o p o u n d e d a similar form o f r e d u c . ., . , ,. , . , . tiomsm, w i t h a m o r e epiphenomenalist emphasis, in his r F B B C lectures The Mind Machine.

All our actions are products of our brains . . . We feel ourselves, usually, to be in control of our actions, but that feeling is itself a product of the brain, whose machinery has been designed, on the basis of its functional utility, by means of natural selection. W e are machines, but machines so wonderfully sophisticated that n o one should count it an insult to be called such a machine. . . . T h e sense of will is an invention of the brain. Like so m u c h of what the brain does, the feeling of choice is a mental model - a plausible account of h o w w e act, which tells us n o more about h o w decisions are really taken in the brain than our perception of the world tells us about the computations involved in deriving it. 4 Such assertions as these - namely, that h u m a n beings are machines, or that the b e h a v i o u r of a h u m a n b e i n g is n o m o r e than the b e h a v i o u r o f their n e r v e cells, or that decisions are taken in and (apparently) b y the brain — are n o t science b u t metaphysics. W h e t h e r such venerable metaphysical pictures are r e n d e r e d any m o r e plausible by m o d e m science than they w e r e in antiquity b y D e m o c r i t u s , Epicurus or Lucretius m o r e than 2,000 years ago is of some interest. Precisely because the various forms of ontological and explanatory

2

Ibid., p. 7. Ibid., p. 8. 4 C. Blakemore, The Mind Machine (BBC Publications, London, 1988), pp. 2 7 0 - 2 . T o be sure, in the sense in which all our actions are products of our brains, so too is all our knowledge (including our knowledge of neuroscience). And just as the fact that the normal functioning of the brain is a necessary condition of knowing anything does not show that we do not know anything, so too the fact that normal functioning of the brain is a necessary condition of acting does not show that we do not act. W e normally feel, i.e. think, that we are in control of our actions, and so indeed we are the fact that this feeling is 'a product of the brain' does not show it to be false. The 'sense of will' is not an invention of the brain, since brains do not invent anything, and the feeling of choice is not a 'mental model', since it is not a model. That we often act because we choose so to act is not a 'plausible account of how we act', since it is not an account of how we act at all. As argued in chapter 3, no decisions are taken in the brain, and our perception of the world is not derived from any computations made by the brain. For perceptions are not derived from anything, and the brain does not compute anything. Are we machines? Only if there are 'machines' that can feel pain and take pleasure in their activities, that have desires and purposes of their own, that are capable of thought and action, that can deliberate on courses of action and decide what to do on the basis of reasons, and that are responsible for what they do. 3

Reductionism

357

reductionism are metaphysical theses c o n c e r n i n g the logic of existential attributions and the logical structures of explanation, they are n o t o p e n to scientific confirmation or disconfirmation. If they are to b e confirmed or confuted, t h e n it will b e by analytical a r g u m e n t . Classical reductionism and unified science

I n order to evaluate such reductionist claims, it is n e c e s , . .. , , sary first to clarity w h a t reductionism is and w h a t forms

it m a y take. In the broadest sense, r e d u c t i o n i s m is t h e c o m m i t m e n t to a single unifying explanation of a type of p h e n o m e n o n . In this sense, Marxism advocates a reductive explanation of history, and psychoanalysis defends a reductive explanation of h u m a n behaviour. M o r e specifically, reductionism in science is a c o m m i t m e n t to the c o m p l e t e explanation of the n a t u r e and behaviour of entities of a given type in terms of the nature and b e h a v i o u r of their constituents. T h e ideal o f ' u n i f i e d science', advocated b y the V i e n n a Circle positivists 5 in the 1920s and 1930s a n d a d o p t e d b y t h e later logical empiricists i n the 1950s, was c o m m i t t e d to w h a t has b e e n called 'classical reductionism'. 6 This c o n c e p t i o n held that the objects of w h i c h the w o r l d consists can b e classified into hierarchies such that t h e objects at each level of classification are c o m p o s e d of objects comprising a l o w e r level. T h e lowest level was conceived to be constituted by the elementary particles investigated by fundamental physics. A b o v e this, i n successive levels, are atoms, molecules, cells, multicellular organisms and social groups. Investigating each level is the task of a given science (or sciences) t h e purpose of w h i c h is to discover the laws that describe the b e h a v i o u r of entities of t h e k i n d i n question. T h e reductivist p r o g r a m m e is to derive the laws of any given level from the different laws describing the b e h a v i o u r of entities at the l o w e r level. Derivational reduction, thus conceived, requires, in addition to the laws at the r e d u c e d a n d r e d u c i n g levels, bridge principles identifying the kinds of objects at the reduced level w i t h specific structures of objects comprising the r e d u c i n g level. 7 R e d u c t i o n i s m , in its classical form, was a bold a n d s w e e p Scientinc reductionism and . , . , , . , . , . , . , . , ... m g thesis a b o u t o n t o l o g y and a b o u t the logical character oy metaphysical materialism ° ° of scientific explanation. It was an e r m n e n d y philosophical thesis, driven by t w o p r i m a r y considerations. T h e first was the apparently successful reduction, in a few domains of science, of fragments of o n e science to elements of another. So, for example, the interactions b e t w e e n stuffs of various kinds is successfully explained in terms of the atomic and valency theories of chemistry. T h e second was a deep c o m m i t m e n t to metaphysical materialism, w h i c h is an ontological doctrine typically p r o p o u n d e d in opposition to Cartesian dualism. In its simplest and warranted form, it 5

See the English translation of the Manifesto of the Circle of 1929: The Scientific Conception of the World: The Vienna Circle (Reidel, Dordrecht, 1973), §2. Rudolf Carnap's hand is evident in the characterization of the reductive 'constitutive system' envisaged (see esp. p. 11). 6 The terminology is derived from John Dupré, The Disorder of Things: The Metaphysical Foundations of the Disunity of Science (Harvard University Press, Boston, MA, 1993), to whose discussion of reductionism in chapters 4-7 we are much indebted. 7 The classical formulation of such classical reductionism is given by P. Oppenheim and H. Putnam in their paper 'The unity of science as a working hypodiesis', in H. Feigl et al. (eds), Minnesota Studies in the Philosophy of Science, vol. 2 (University of Minnesota Press, Minneapolis, 1958). The flaws in this conception are well discussed by Dupré.

358

On Method

amounts to a denial that there are mental or spiritual substances. In its simplest and crudest form, it involves the claim that everything that exists is material. In this form, it claims that the mind is the brain (hence the proliferation in recent years, in the wake of Noam Chomsky, of the misconceived phrase 'the mind/brain'). In less simple and crude form, it is the claim that mental states, events and processes are in fact neural states, events and processes, that mental attributes are in fact identical with neural ones. ,. , , Ontological materialism has little to be said for it. Denial Untoloaical materialism , , . . . . . , that there are mental or spiritual substances does not imply that the only things that exist are material objects (and material stufis). For evidently laws and legal systems, numbers and theorems, games and plays, are neither material objects nor stufis. Indeed, the colours, lengths and weights of material objects, not to mention their capacities and dispositions, are not themselves material things, although it makes perfectly good sense to speak of there being such properties as colours, lengths and weights, and such dispositions as solubility and elasticity. More importantly, wars, revolutions and cultures, performances of plays, birthday parties and funerals, are not material objects - but there are such things, they occur, happen, or exist at a time or for a time. One might modify the claim: everything there is, one might suggest, is made of, or consists of, matter. But this is just as misconceived, since laws and legal systems, numbers and theorems, games and plays, political parties, a society and its culture, inflation and economic growth, are not made of matter and do not consist of matter. Moreover, denial that there are immaterial substances does not imply that the only thing relevant to the explanation of the properties and/or behaviour of things that do exist - indeed, even of material things that exist — is the matter of which they are made. Organs and artefacts are explained primarily by reference to their function, not merely by reference to their material constitution. The behaviour of sentient creatures in general is explained partly in terms of their goals, and of human beings in particular also in terms of reasons and motives, not in terms of the material of which they consist. Even more obviously, the explanation of events and processes such as Hannibal's victory at Cannae, or the decline of the Roman Empire, the Industrial Revolution, or the rise of Romanticism has nothing to do with the matter of which the explananda are made, since they are not made of anything. So, the materialist might venture a much more modest claim: everything that is made of anything is made of matter. Certainly we should concede that minds are not made of immaterial substance8 — but then, if the argument of §3.10 was correct, minds are not made of anything, and all talk of the mind is a mere fagon de parler for talk of specific human capacities of thought, memory and will and of the exercise of those capacities.9 Legal B

Of course, Descartes never claimed they were. He held that they are immaterial substances, not that they are made of 'immaterial stuff'. 9 To be sure, if that argument was correct, then Crick's claim that our minds are 'the behaviour of our brains' is no more coherent than the idea that our sakes are the behaviour of our friends who do things for our sake. Nor, more circumspectly, is making up my mind, changing my mind, or being in two minds whether to V the behaviour of my brain, even though the possibility of my making up my mind, changing my mind and being in two minds whether to V\% possible only if my brain is functioning normally in appropriate respects.

Reductionism

359

systems consist of laws and not of matter; poems consist of stanzas, not of ink; and revolutions consist of human actions and events. The materialist might grant that this is what laws, and poems, and revolutions consist of, but deny that they are made of anything. We can concede this too. But even if it is true that everything that is made of anything is made of matter, this thesis goes no way to sustain any form of ontological reduction according to which all 'entities' are reducible to material entities. Nor does it support any form of explanatory reduction according to which the properties and behaviour of everything that exists are to be explained in terms of the properties and behaviour of its constituent matter. , . , . „ TI Human beings are not. ontologically .-,,,,,. t reducible to their nervous systems

That everything that is made of anything is made of . 1 , 1 1 matter does not show that human beings are ontologically & . &"- ' c a n e x P l a m w h a t p0U,ers neural connections must obtain and what neural activities must take place in order for it to be possible for the animal to possess and exercise the powers it naturally possesses. In the case of human beings in particular, neuroscience may aspire to explain the neural conditions for the possibility of the mastery of a language, the possession of which is itself a condition of the possibility of rationality in both thought and action. However, neuroscience cannot

Reductionism

365

displace or undermine the explanatory force of the good reasons we sincerely give for our behaviour, or invalidate the justifications we give for rational behaviour. The rationality of behaviour that is motivated by good reasons is not given a deeper explanation by specifying the neural facts that make it possible for creatures such as us to act for such reasons. When we apprehend the propriety, adequacy or goodness of the reasons for which a person acted, then we fully understand why he did what he did. What neuroscience can do, however, is contribute to the Neuroscience can contribute to „. . , , , 1 . t T , . r• • i • explanation or irrational or partly irrational action. It may r J explanations of irrational action , , be able to explain why a person is more prone than norma] to certain mental states - for example, of depression, which makes him more Hable to act for a certain kind of a reason than someone who is not thus depressed. This can play an important role in explaining human behaviour in certain circumstances. But it is also important to note that such an explanation need not supersede the reasons the depressed person might give for committing suicide, for example. That he is depressive, perhaps pathologically depressive, does not imply that his reasons for killing himself are mere rationalizations, which play no role in rendering his behaviour intelligible. The neuroscientific explanation may complement the explanation the agent offers in terms of reasons, without rendering those reasons irrelevant. Furthermore, neuroscience can explain — indeed, specialNeurosaence can [explain forms of , . . , , , . , , n . . . izes in explaining - how gross pathological deficiencies incapacitation . . ° ,, T,~ in the exercise of normal human capacities result from damage to the brain. So it can brilliantly explain why patients cannot- behave as normal humans can in a multitude of different ways. In particular, it may explain why such patients are, in one way or another, incapable of acting rationally in certain respects. The remarkable neuroscientific successes in explaining Neuroscience cannot explain . .. , „ , . . . ,,.,.,.. , . , , . certain kinds of psychological pronenesses and liabilities normal human behaviour , of temperament, and in explaining pathological behaviour and deficiencies, do not demonstrate that neuroscience can or should ever aspire to explain normal human behaviour (as opposed to explaining the neural conditions of its possibility). What neuroscience can do is to explain, for normal human beings, how it is possible for them to be open to reason. But it cannot explain the rationale of human actions in the particular case, or elucidate what makes a certain reason a good reason, it can identify necessary conditions for the exercise of human capacities. But it does not follow that it is, or ever will be, in the position to specify a set of neural conditions that are sufficient conditions for characteristic human action in the circumstances of life. To explain typical human behaviour, one must operate at the higher, irreducible level of normal descriptions of human actions and their various forms of explanation and justification in terms of reasons and motives (as well as causes). These descriptions will cite multitudinous factors: past and prospective events that in given circumstances may constitute the agent's reasons for action; the agent's desires, intentions, goals and purposes; his tendencies, habits and customs; and the moral and social norms to which he conforms. Our common or garden explanations of human conduct focus upon what it is that human beings do — typically identified in terms oí action rather than movement (as noted in §8.1, we want to understand a person's moves, not his movements). Such identifications

366

On Method

are highly context- and circumstance-dependent. Having clarified what it was that a person did or was doing in a given circumstance, it may still not be obvious why he acted as he did. This in turn may be explained by reference to his intentions, goals and purposes his reasons or motives, his habits, customs and inclinations — in the context of human social, moral and legal life. Neuroscientific explanation is not in competition (let alone in conflict) with these kinds of explanation; but neither does it reduce these forms of explanation to neuroscientific ones.

13.2 Reduction by Elimination Derivational reductionism looks singularly unpromising.15 „ , , , , , ° But over the last twenty years there has been a more radical suggestion. Conceiving of our common or garden explanations of human conduct as parts of what they contemptuously refer to as 'folk psychology', some American philosophers, most notably Steven Stitch and Paul and Patricia Churchland, have suggested that this 'theory' of human behaviour is destined to be eliminated by a future neuroscientific theory. So they in effect advocate the eliminative reduction of psychological explanation of human behaviour to a future neuroscientific theory that will explain all human conduct. They accept the fact that the 'entities' that are involved in psychological explanation (reasons, motives, beliefs, desires, etc.) are not reducible to neuroscientific structures. But that, they aver, is because such 'entities' are mere fictions. The psychological level of explanation that intervenes between behaviour and neuroscientific theory is bogus, and requires elimination. All human behaviour will be fully explained by neuroscience without recourse to the primitive terms of psychology - either of the 'folk' variety or of the experimental sort. , „ ,, „ , , Folk psychology, the eliminativists claim, incorporates What folk psychology is , ,. . . , ,, .. , , the ordinary conceptual rframework that we all use m supposed to be , , ,, , - , order to comprehend, predict, explain and manipulate the behaviour of humans'. This framework 'includes concepts such as belief, desire, pain, pleasure, love, hate, joy, fear, suspicion, memory, recognition, anger, sympathy, intention, and so forth'. Accordingly, it constitutes our conception of what a person is. But eliminativists hold that folk psychology is much more than a mere array of concepts. It is also a theory about human behaviour. For 'the relevant framework is speculative, systematic, and corrigible,... it embodies generalized information, a n d . . . it permits explanation and prediction in the fashion of any theoretical framework'. 16 Being a theory, it propounds causal, explanatory laws (e.g. a person who incurs severe bodily damage will suffer pain; a Eliminative reductionism

15

The following section is a much shortened version of P. M. S. Hacker, 'Eliminative materialism', in S. Schroeder (ed.), Wittgenstein and Contemporary Philosophy of Mind (Routledge, London, 2001), pp. 60-84. 16 P. M. Churchland, 'Folk psychology', in S. Guttenplan (ed.), A Companion to the Philosophy of Mind (Blackwell, Oxford, 1994), p. 308.

Reductionism

367

person who suffers pain will wince; a person denied food for any length of time will feel hungry), and it warrants predictions (e.g. people who desire thatj?, and believe that K'ing will bring it about that p, and have no overriding desires or preferred strategies, will generally try to V). 'Folk psychology', thus conceived, is held to be a ther Three alleged failings of 'folk , , , , . , tr M , ,_ „ or . . , y that belongs together with folk astronomy, folk " * physics', 'folk thermodynamics', 'folk biology', and so forth. The latter are held to be rudimentary or crude theories that have, over the last four centuries, been replaced by serious scientific theories. Folk psychology is destined for a similar fate. For it stands accused on three counts, (i) Explanatory, predictive and manipulative failures: folk psychology does not explain what sleep is or why we need it, how learning transforms us from infant to educated adult, the grounds of intelligence, how memory works, what mental illness is, or how it is to be cured, (ii) It has not progressed significantly in the last 2,500 years, has not shown the expansion and developmental fertility expected of a successful theory, (iii) It cannot be smoothly integrated into the emerging synthesis of the physical, chemical, biological and neuro-computational sciences. In particular, it is admitted, there is Htde prospect of a smooth theoretical reduction of the concepts, entities and laws of folk psychology to the more basic concepts, entities and laws of the advanced sciences of neurobiology, chemistry and physics.17 The first two counts suggest that folk psychology should The alleged faculty of its concepts , , , , , ,. , , , , r ,, , , . be replaced by the science of experimental psychology, r suggests that folk psychology requires . . . . . , ,. . ,. mst as, according to the ehmmativists, physical astronomy elimination

J

>

&

r

;

i

replaced folk astronomy, thermodynamics replaced folk thermodynamics, and biology replaced folk biology. But this is not what the eliminativists have in mind. In their view, the concepts of folk psychology are vacuous - comparable to the concepts of phlogiston, caloric, crystalline heavenly spheres and élan vital. So folk psychology must be eliminated. For, 'once folk psychology is held at arm's length and evaluated for theoretical strength in the way that any theory is evaluated, the more folkishly inept, soft and narrow it seems to be'. 18 Its concepts are as vacuous as phlogiston, caloric, witch and other similarly empty concepts that have been discarded with the progress of science. Hence, too, empirical psychology, which employs much the same array of vacuous concepts, will hkewise be eliminated by the march of neuroscience. The crude laws of folk psychology and the alleged laws of experimental psychology will be replaced by precise laws of a future neuroscientific psychology.

13.2.1 Are our ordinary psychological concepts theoretical? tIt,

, , , ,. Our common explanations of human behaviour why it is thought that our ordinary . . _ , . . . ,_ , , , . , , .,, ,. , a rich array of psychological concepts. But why psychological concepts are theoretical i n , o r anyone suppose that they involve any theory of 17

invoke , ,, should J human

Ibid., pp. 310f P. S. Churchland, Neurophilosophy: Toward a Unified Science of the Mind/Brain (MIT Press, Cambridge, MA, 1986), p. 395. 8

368

On Method

behaviour? One reason given is that all judgement involves the application of concepts, that every concept is a node in a network of contrasting concepts, and that a concept's meaning is fixed by its position in the network. But any network of concepts is a speculative assumption or theory, minimally a theory as to the classes into which nature divides herself and the relations between them. 19 Learning the ordinary psychological vocabulary involves learning appropriate generalizations that specify conditions of correct application. Psychological terms are implicitly defined by such generalizations, and they are part of the theory of folk psychology. And our ordinary psychological explanations of human conduct include generalizations, which is a characteristic feature of theory. , This is a misconception. It is correct that concepts are Logical relationships do not imply . , , , . ,. .,.,. r r , .. ' interrelated by way ofr implication, compatibility and r í theoretiaty ' ' . incompatibility. But this does not imply that ail concepts are theoretical. That something is red all over implies that it is not blue, green, yellow, etc. all over, that it is darker than any pink object, and more like an orange object in respect of colour than like a yellow one. These propositions are not theoretical but conceptual or grammatical truths that are partly constitutive of the meaning of the word 'red'. That a piece in chess is the king implies that it is the piece that can be checked, that it moves one square at a time, that it can castle, etc. These propositions do not imply that 'chess king' is a theoretical concept, although, of course, they partially define what a chess king is. , Every concept is indeed embedded in a ramifying netThe holistic character of language , _, . , -r a, ^ - . . , , :. A A work of r concepts. This does not signify theoretiaty, but does not imply theoretiaty ,, , , a T 1 the normative and holistic character of language, it shows that an expression is to be used in accordance with the rules that determine its meaning. And it shows that an expression has a meaning only as an expression in a language, in co-ordination with a host of further expressions with which it is conceptually or grammatically related by rules stipulating compatibilities, incompatibilities and implications. But it would be absurd to claim that all concepts are theoretical. For that would render vacuous the claim that a given concept - for example, of a meson or a quark - is theoretical, by contrast with, say, the concepts of a tree, a game of cricket or a table. It is mistaken to suppose that a network of concepts is A conceptual network does not , . . , , , , t , , ? . - • , a speculative assumption or theory , let alone a theory as r r imply theoretiaty , , . , . . to the classes into which nature divides herself. Nature does not 'divide herself into anything. Our concepts have various purposes. In so far as the purpose of a set of concepts is the scientifically and theoretically fruitful classification of natural phenomena, the concepts we introduce may be more or less useful relative to that purpose. But it is we who classify things thus, not nature. And how we classify things is determined not by nature but by our theoretical interests. Even scientific classification does not yield absolute, precise, purpose-independent categories, determined by the natural order of things. In biology, morphological criteria often quarrel with evolutionary criteria, and neither uniformly deliver determinate answers - assigning organisms to species, for example, is no less purpose-relative, variable and partly arbitrary than common or garden 19

P. M Churchland, Matter and Consciousness, rev. edn (MIT Press, Cambridge, MA, 1988), p. 80.

Reductionism

369

classification. There are many different ways of classifying the products of evolutionary processes, and whether one way or another is the most fruitful depends upon the specific purposes we have and on the peculiarities of the organisms in question. ., , i t However, classifying for the purpose of scientific theory Not all classification ts done tor . . . , c , .c is only one kind or ciassihcatory purpose. Our various f , " classifications of artefacts {tools and weapons, clothing and buildings, paintings and sculptures), of their merits and demerits, and of the skills needed to make them are by and large not for scientific purposes at all. But even natural, as opposed to artefactual, objects may be classified for purposes other than scientific ones. They may be classified relative to the varieties of purposes and interests that we have in them. (The concept of a tree is one that finds no place in systematic botanical classification, but is a highly useful one for a multitude of other human purposes.) Finally, a vocabulary is not a theory. The specialized A vocabulary is not a theory , , , c r . c . c ' vocabularies of the arts ana crafts of our culture, of games and of rituals, of property and possession, of morality and law, are not theories of anything. English contains no theory, although by now it contains a host of theoretical terms that have accrued over the past few centuries as the theoretical sciences have evolved. A language is not a theory of anything, although it may provide the resources for the articulation of indefinitely many theories, including mutually contradictory ones. The Ptolomaic, Copernican and Keplerian theories of the solar system are all formulable in English and are mutually incompatible, but that does not make English an inconsistent language. O f course, all articulate judgement involves the applicaThat observation is concept-laden . . . . . , , , , . . . , , tion or cconcepts.r Articulate observation is concept-laden. does not imply that it is theory-laden . . . , , , But it does not follow that it is therefore theory-laden. There must be a contrast between what is theoretical and what is not if the term 'theoretical' is to have any content. A scientist's description of particle decay in a cloud chamber will be theory-laden, involving the use of non-observational theoretical terms. But a description of a garden as tidy, with daffodils and tulips in bloom, is not. Neither is the description of a person as wondering whether to go to the theatre tonight, as thinking that it would be enjoyable, and deciding to get tickets. In chapters 4—8 we surveyed a range of fundamental Acquiring concepts is not the same 1 1 - 1 „, , 1 , . , psychological concepts, w e noted, in many cases, the as teaming a theory . , , various ways in which these concepts might be acquired. Some, we suggested, are grafted onto preHnguistic natural expressive behaviour (e.g. 'hurts', 'want'), while others (e.g. 'know', 'believe') are grafted onto pre-existent linguistic behaviour. What should be evident is that in acquiring such concepts a child is not learning a theory of anything; he is learning forms of human behaviour. In learning to replace his cries of pain with 'It hurts' and 'I have a pain', in learning to replace his cries of frustrated endeavour to get something with an 'I want', in learning to herald an action with an 'I am going to', the child is not learning a folk theory of human conduct; he is learning human conduct, learning to give expression to his pain and to express his desires and intentions. He is learning the Janus-faced use of psychological terms - their expressive (and later reportive) use in the first-person present and their descriptive use in the third

On Method

370

person. And in learning the rudimentary use of 'I know', 'I believe', and 'I think', the child is not learning a theory about his own inner states, but a language-game which presupposes a grasp of the difference between a well-founded assertion and one that is not adequately founded, or between expressing one's own judgement and expressing a derived one. The uses of psychological predicates that the child learns The uses of psychological predicates , , , „. r , , . -,t i ., are far removed from theoretical concerns. Hrst-person learnt by a child are not theoretical . . . . . expressions or sensation, desire, intention, etc. are not hypotheses about one's own mental states. Ascriptions of psychological attributes to other people are not hypotheses about the existence of inner states in accordance with a common-sense theory. For thinking, believing, hoping and fearing that p are not inner states of a person hidden behind his behaviour. His behaviour, including his avowals of thought, belief, hope and fear, is not inductive evidence, let alone evidence for postulating the existence of unobserved entities. It is no theory that an utterance of the form 'I Fthat p' (where CV is a verb such as 'think', 'believe', 'fear', etc.) is a criterion for the speaker's King. It is, rather, as we have seen in §3.3, a rule for the use of the verb that is partly constitutive of its meaning. Psychological concepts are not concepts of imperceptPsychological concepts are not . , , . . , . , r , 1 i •• ihle entities, like genes or viruses, or concepts or tneoretr , concepts of theoretical entities , ,-, , -, ical entities, hke mesons or quarks. They are not concepts of'entities' at all. Our concepts of beliefs, thoughts, hopes, fears, expectations, etc. are not concepts of kinds of things, but abstractions from believíngs, thinkings, hopings, fearings and expectings. In one sense, what many of these verbs signify is often perfecdy observable. For it is a conceptual confusion to suppose that the evidence for someone's suffering, joy or grief, for his believing or thinking, fearing or hoping, consists of'bare bodily behaviour', of mere physical movements. O n the contrary, we can typically see another's rage (when he is raging), see the grief or anguish visible upon his face and in his demeanour (when he is grieving or in anguish). One can hear a person's thoughts when he tells them to one, and read them if he writes them down. One needs no theory, folkish or otherwise, to hear or read the expressed thoughts of another. The fact that we need not reveal our thoughts, that we can sometimes suppress any manifestation of feelings, and that we can sometimes pretend, he and deceive does not make psychological predicates theoretical.

13.2.2

Are everyday generalizations about human psychology laws of a theory?

, ,, , , , , Eliminativists contend that folk psychology incorporates The alleged theoretical doctrines of , . , ,r „ , , . not merely a large arrayJ of theoretical concepts, but also folk psychology \ ° , a large body of theoretical doctnne. They hold that the rudimentary, allegedly causal laws of folk psychology are used to explain and predict human behaviour in the standard 'covering-law' fashion. But this is confused. We must distinguish various kinds of proposition here. Three types of generalization _. .. . ... 1T1 . , , cc First, propositions such as (1) People who surfer a sharp pain tend to wince' or (2) 'Angry people tend to get impatient' do not state causal laws.

Reductionism

371

Wincing is a logical criterion for being in pain - it is a form of pain-behaviour. That people commonly wince when in pain is not a discovery resulting from correlating people's winces with their being in pain, since we identify people's being in pain by reference to (among other things) their wincings. Similarly, although it is true that angry people tend to be impatient, this generalization is not a causal law, but a characterization of one form which anger may take. The anger is not a cause of the impatience; rather, it manifests itself in impatience. Moreover, to explain A's impatience by reference to his anger is no more to give a nomological explanation than is explaining someone's weeping by reference to their grief It is not like explaining someone's white hair by reference to their age. Second, propositions such as (3) 'People who want it to be the case that p and who believe that K'ing will bring it about that p, and have no overriding reason not to V, will generally V or try to V and (4) 'People who believe that p, doubt whatever they apprehend as being incompatible with p' are likewise not simple causal generalizations. Rather, they are generalizations that rest on conceptual connections and are partly constitutive of the meanings of their constituent terms. If someone claims to want to bring it about that p, believes that K'ing will bring it about that p, and has no reason not to V, and yet does not V, then there is a reason for thinking that he does not really want to bring it about that p. If one knows or believes that if it is true that q, then it is false that p, and if one knows that p, then of course, one will not normally believe that q. But that is no empirical generalization. For not doubting that q here would be a logical criterion, for one's not knowing that p, or of one's not knowing that if q then not p, or of not grasping the inference rule, and hence of not understanding the conditional. To grasp such truths as (3) and (4) is not to know or believe any psychological laws or empirical generalizations, but rather to have grasped the concepts of knowledge, belief, doubt and want. Third, propositions such as (5) 'A person denied food for any length of time will feel hungry' and (6) 'Injury usually causes pain' are equally dubious candidates for causal laws allegedly discovered by the bogus 'folk theory' of human psychology. For injury is a circumstance in which pain-behaviour is a criterion of pain, and deprivation of food is a circumstance in which desire for food is a manifestation of hunger (rather than greed, for example). Injury is not merely causally, but also conceptually, linked with pain, as food deprivation is not merely causally, but also conceptually, linked with hunger. Finally, one can indeed give examples of something that Genuine examples of folk . , , , , LC „ , , , ™, • , > might be deemed folk psychology . They are propositions such as (7) 'Spare the rod and spoil the child' or (8) 'Once bitten, twice shy'. These can hardly be taken to be theoretical statements, and whatever psychological terms occur in such statements are not implicitly defined by such generalizations. it is, of course, true that we explain and predict people's Psychological explanation and pre- , , , , . , , , , - , r r ,. . , , behaviour by reference to their thoughts, beliefs, wants, diction does not conform to the . . ° j ,, intentions, likings and dishkinss. But it is mistaken to & & covering-law model ' suppose that such explanations and predictions generally, or even typically, conform to the subsumption-theoretic covering-law model. O n the contrary, they typically rely on principles of practical reasoning. That people generally do

On Method

372

what they take themselves to have good reason to do, that they try to execute their intentions, that they tend to pursue what they see as desirable, are not causal generalizations Even when appropriate generalizations and predictions do conform to the covering-law model, they are not in any sense proto-scientific laws. 'He hasn't eaten for hours, so he must be hungry' is no more theoretical or proto-scientific than 'It is starting to rain, so the laundry will get wet'. We must obviously concede that our common or garden explanations are sometimes mistaken. Our self-understanding may be defective, and we sometimes deceive ourselves so the explanations we offer of our own behaviour can be wrong. They may be correctable by others. But this does not suggest that our ordinary psychological concepts are obsolete and need to be replaced; it only shows what should be obvious: namely, that one is not always the final authority on one's own emotions and motives. Equally obviously, we can err in the explanations we give of the behaviour of others, and sometimes their behaviour and motivation may be altogether opaque. But this does not indicate deficiency of theory, let alone that the positing of unobservable entities such as beliefs and desires is unwarranted or ineffectual.2(l For ascribing knowledge, belief, desires or intention to others is not positing anything, and knowledge, belief, desire and intention are not theoretical entities of any kind. What it does indicate is that the logical criteria for the ascription of psychological attributes to others are, as we argued in §3.3, defeasible. It shows that human beings are fallible, and also that there is a degree of indeterminacy about the mental {e.g. about motivation, or about the authenticity of emotion). Various aspects of human behaviour can undeniably be clarified by empirical psychological research. This augments our understanding, but it supplements and does not displace our ordinary explanations of human behaviour. And the same applies to neuroscientific insights into the causes of a wide variety of behavioural deficiencies.

13.2.3 Eliminating all that is human As we saw above (§13.2), eliminativists offer three reasons , c „ , , , ,,, ... . , , , . why folk psychology should be ehmmatively reduced m favour of future neuroscience. The first was its explanatory and predictive failures in, for example, not explaining what sleep is, how learning is effected, the grounds of intelligence, or what mental illness is, and how it is to be cured. But since our normal psychological vocabulary is not theoretical and our normal observations, explanations and generalizations are not laws of a theory of any kind, the objection falls fiat. Empirical psychology should and does offer theories and explanations of why we need sleep, what the basis of differential IQ is, what different kinds of mental illnesses there are, and how they can be cured. But it is as absurd to castigate our ordinary conceptual framework for failing to deliver such goods as it would be to blame our common or garden vocabulary of sticks and stones, chairs and tables (and attendant The three reasons for elimination . ,

20

Cf. P. M. Churchland, Scientific Realism and the Plasticity of the Mind (Cambridge University Press, Cambridge, 1979), p. 91.

Reductionism

373

h u m d r u m generalizations), for failing to c o m e up w i t h theories of matter and laws of mechanics. T h e second reason was that so called folk psychology has n o t progressed over the last 2,500 years. Progress m i g h t i n d e e d b e a r g u e d to b e t h e f o r m o f science. B u t since o u r psychological language is n o t a theoretical, scientific language, the accusation of 'lack of progress' is misguided. For it is less than clear w h a t , in this context, 'progress' m i g h t m e a n . O n e m a y argue that the empirical science of psychology has i n d e e d progressed — that m u c h k n o w l e d g e has b e e n attained a n d that old theories have b e e n refuted a n d replaced by m o r e adequate theories. B u t this has n o bearing o n o u r c o m m o n psychological v o c a b u lary, for, as has b e e n argued, a vocabulary and the n e t w o r k of logical relations b e t w e e n its constituents are n o theory, a n d t h e explanations w e give of o u r o w n b e h a v i o u r and of the b e h a v i o u r of others are n o t theoretical. O n e m i g h t i n d e e d argue that the English psychological vocabulary (like t h e English aesthetic vocabulary) has b e e n enriched o v e r the past thousand years, b u t that w o u l d b e ill-characterized as a form of scientific progress. T h e third reason was that ordinary explanations of b e h a v i o u r cannot b e integrated into 'the e m e r g i n g synthesis of the physical, chemical, biological and n e u r o - c o m p u t a t i o n a l sciences'. 2 ' Since there can b e n o bridge principles linking psychological explanations w i t h neuroscientific ones, a n d since there are n o strict psychological laws, the question of the derivational reduction of ordinary psychological explanations to neuroscientific ones cannot arise. H o w e v e r , this gives n o support to eliminative reduction. R a t h e r , it indicates the absurdity of any form of r e d u c t i o n of the psychological, n o matter w h e t h e r derivational or eliminative. U n d e r l y i n g these three considerations is a primitive m o d . ,. r , . . , ,,, , . e r n belief, characteristic of o u r times, that all k n o w l e d g e . , and ail g e n u i n e understanding are scientific. As R i c h a r d & . . . D a w k i n s succinctly puts it: 'Science is the only w a y w e k n o w to understand the real world.' 2 2 'Science', in such declarations of faith, is physical science — in particular, microbiology, chemistry and, ultimately, physics. For, it is argued, 'in t h e d i m e n s i o n of describing a n d explaining the world, science is the measure of all things, of w h a t is that it is a n d of w h a t is n o t that it is not'. 2 3 , B u t , first, there is n o such thing as 'explaining the w o r l d ' , There is no such thing as ivrr c i • • i-cr i , , , , ,,, only 3 different ways 01 explaining different p h e n o m e n a in explaining the world ' . . . . , the w o r l d . T h e theories of the various natural sciences d o n o t , and d o n o t p u r p o r t to, describe a n d explain everything describable a n d explicable. Law, e c o n o m i c s a n d sociology describe a n d explain legal, e c o n o m i c and sociological p h e n o m e n a n o less than physics describes a n d explains physical p h e n o m e n a and chemistry describes and explains chemical p h e n o m e n a . W i t h i n their p r o p e r d o m a i n s , the social That all knowledge and genuine ,. .n . . understanding are scientific is a pn, , ,. e mitwe modern belief

21

P. M. Churchland, 'Folk psychology', p. 311. R . Dawkins, 'Thoughts for the millennium: Richard Dawkins', in Microsoft Encarta Encyclopaedia 2000 (Microsoft Corporation, 1993-9). 23 W. Sellars, 'Empiricism and philosophy of mind', in his Science, Perception and Reality (Routledge and Kegan Paul, London, 1963), p. 173. 22

374

On Method

sciences are no less a measure of what is and what is not. And history, which is neither a natural nor a social science, is a measure of what was that it was and of what was not that it was not. Moreover, there is no prospect whatsoever that legal, economic sociological and historical phenomena should be explained by, let alone be reducible to any natural or biological science. It is grotesque to suggest that these subjects are all pseudo-sciences or mere fables, enmeshed in a vacuous and obsolete vocabulary. Is it being suggested that our ideas of how to explain the fall of the Roman Empire, the rise of Protestantism, the outbreak of the French Revolution, the causes of the First World War, are all fictions? „ . „ Secondly, it is absurd to suppose that science, no matter Science is not the measure of all . . . , , . , , r ,. whether social or natural, is the primary measure of what things ' . ? i does and what does not exist. One needs no science to discover or come to know that there is a tree in the garden or that there are no trees in one's room. Nor does one need any science to explain that one went to Paris because one promised a friend to be there. Not everything that can be known can be known by mere observation, but nothing at all could be known without mere observation - and the ability to learn facts about the world around us is a prerogative of any human being, antecedent to science and the acquisition of scientific knowledge. It is simply false to suggest that all observation is theory-laden. Thirdly, it is wrong-headed to suppose that the only There are forms of explanation and r , ,. .r 1 1 , . r , ,. . ., . forms of understanding are scientific, and that the only understanding that are neither saent° . ' ifir nor theoretic I respectable forms of explanation of empirical phenomena are theoretical. It is wrong to suppose that philosophical understanding and philosophical explanation of conceptual error and confusion are modelled on scientific modes of understanding natural phenomena. It is no less mistaken to suppose that historical understanding is modelled on the understanding that characterizes physics, chemistry or neuroscience. Only dogmatism can lead one to suppose that either there is no such thing as understanding aesthetic phenomena - understanding works of literature, music, painting, sculpture and architecture — or that such understanding apes the understanding that scientists hope to achieve of physical or chemical phenomena. And we are all indebted to the understanding of human nature that is displayed in the works of such simple folk as Tolstoy and Dostoevsky, Proust and Henry James. Scientific theories are replaced in the course of scientific Abandoning theories and abann i 1 xr 1 , . progress - Ftolomaic astronomy by Newtonian physics, doning concepts , , , i and that in turn by relativity theory; caloric theory by thermodynamics. Sometimes concepts are jettisoned as vacuous {e.g. phlogiston, caloric, élan vital) and replaced by more fruitful ones. But fundamental changes in astronomy did not lead to the abandonment of the concepts of sun, moon and stars, any more than changes in chemical theory led to the abandonment of the concepts of burning and rusting, or of heat and cold, or changes in the life sciences led to jettisoning the concepts of a living creature, a dead creature and. an inanimate thing. As we have seen, the eliminativists construe our psychological concepts as proto-scientific concepts on the same level as phlogiston, caloric or élan vital But this is mistaken. Unlike such concepts, our normal psychological concepts are not postulates of any theory, but constitutive elements

Reductionism

375

of the human form of life. Phlogiston was postulated to explain combustion, caloric to explain the transfer of heat, and élan vital to explain life. In due course it became clear that there are no such things — the concepts are vacuous and the explanations mistaken. But, as we have argued, our psychological concepts are not in the least like this. Our psychological concepts are not theoretical concepts Our ordinary psychological concepts , . , c . .c , , . . . , , . , devised tor scientific purposes, though it is true that the are neither theoretical nor empty , , - , , i i - i • 1 psychological and neuropsychological sciences use them, just as chemistry uses the concepts of water and iron, and biology uses such concepts as cat and dog. They are used, among other things, to describe the phenomena that are the subject matter of empirical psychology (but they are also used to exhibit or manifest these phenomena, to give them articulate expression). It may well be that there are technical concepts in current empirical psychology that will be abandoned by future psychology. But this does not show that there are no beliefs and thoughts, perceptions and sensations, desires and intentions. Nothing answers to the concepts of phlogiston, caloric or élan vital. But it cannot be said that nothing answers to our ordinary psychological concepts. There are criteria for the application of these terms, and these criteria are satisfied daily, innumerable times, in the life of every normal human being. It is only when the logic of these expressions is misconstrued, as it is by the eliminativists, and wrongly taken to signify theoretical, unobservable entities, that it can confusingly seem as if it might be that nothing answers to these concepts. Eliminativists construe our ordinary psychological vocabuOn misconstruing the role of our , ^ b d n u s e d ^ Q r d e r tQ c o m p r e h e n d ) p r e d i c t i psychological vocabulary . . , . , i_ i_ i_ e neuroscientists.

14.3 From Nonsense to Sense: The Proper Description of the Results of Commissurotomy „ , , The descriptions given by . . neuroscientists

It has become customary for neuroscientists to discuss , , , - - - , , r the deficiencies that result trom commissurotomy in ccniii ' that involve treating each hemisphere of the brain «is it Jt were a possible subject of psychological attributes. The fascinating work on this subject lia, ••• , tned to show how this can be done. Philosophy can point out when the bounds of sense are transgressed — as we have done with respect to the rnereological fallacy in neuroscience. It can make clear when the conceptual framework which informs a neuroscientist's research has been twisted or distorted. So, it can clarify as we have done — what is awry with the thought that perception involves seeing or having images or that perception is the hypothesis formation of the brain. It can warn - as we have tried to do - against the confusion of supposing that memory is the re-enactment of past experience or is always of the past. It can elucidate why conditioned reactions are not forms of memory and why it is confused to think that memories can be stored in the brain. It can show — as we have shown - why the study of emotional perturbations is not the same as the study of emotions, why emotions are not somatic responses to mental images, and why feelings of emotions are not the awareness of somatic changes that inform us about our visceral and musculoskeletal state. It can explain - as we have explained why mental images are not ethereal pictures and why they cannot be rotated in mental space. And so on. Far from being irrelevant to the goals of neuroscience, the conceptual clarifications of philosophical analysis are indispensable for their achievement.

14.4.2 What neuroscience can and what it cannot do It is perfectly correct, as Crick urges, that it is hopeless to Why neurosaence cannot contrib, .. , .. . . , , . . .,, ... try to solve empirical problems concerning: ¿ the nature ute to the solution of philosophical ,, of consciousness by general philosophical arguments. But, equally, it is hopeless to try to solve conceptual problems concerning the nature of consciousness by empirical methods. PET and f M R I can scan brains, but not concepts and their articulations. Neuroscience can investigate synaptic connections but not conceptual ones. It is not altogether clear what Edelman means by 'a biologically based theory of the mind'. But if neuroscientists produce something that can be so denominated, it cannot be expected to give philosophy 'a new lease on life'. For, in the first place, philosophy has not lost its old lease, which is unlikely to run out as long as mankind continues to think and human beings, including scientists, continue to err for purely conceptual reasons. And in the second place, a new neuroscientific theory of anything cannot give philosophy a new lease or turn, but only a new array of conceptual puzzles to resolve and knots to disentangle. Neuroscientific discoveries (e.g. blind-sight) may pose new conceptual

406

On

Method

p r o b l e m s - they m a y p r o v i d e grist for philosophical mills, b u t n o t solutions for philosophical p r o b l e m s . . „ Why epistemology could not be , , . . grounded in neuroscience

T h e suggestion that epistemology should b e g r o u n d e d in , 1 1 T neuroscience can be p r o p o s e d only by s o m e o n e with anu

infirm grasp of w h a t epistemology is. It is, after all, not an empirical enquiry i n t o h o w , as a matter of fact, h u m a n beings can and d o acquire w h a t e v e r k n o w l e d g e they have - that is learning t h e o r y , w h i c h is a b r a n c h of psychology R a t h e r , epistemology is an a priori enquiry i n t o the w e b of epistemic concepts that is formed b y the c o n n e c t i o n s , compatibilities and incompatibilities b e t w e e n the concepts of k n o w l e d g e , belief, conviction, suspicion, supposition, conjecture, doubt, certainty, memory, evidence and self-evidence, truth and falsehood, probability, reasons a n d reasoning, etc. T h e relevant connections are logical or conceptual — a n d neuroscientific investigations can shed n o light u p o n the normative connections of logic (construing 'logic' broadly). E p i stemology is also c o n c e r n e d w i t h the logical character of justifications of k n o w l e d g e claims, of confirmation and discorrfrrmation, of the differences b e t w e e n deductive and inductive support, of w h a t counts as evident and w h a t stands in n e e d of evidence, and so forth. This t o o is n o t an empirical investigation. It could n o t possibly b e furthered b y t h e discovery of facts a b o u t the brain. It could n o t b e the task o f neuroscience to investigate Why neuroscience cannot investigate . , .. . , , . , . ' . r, , J t h e conceptual n a t u r e of k n o w l e d g e , a n d its relation to the conceptual nature of knowledge: . . . . . . ,. ., , . , , _, , . ry ¡., r . belief. If it w e r e to shoulder this b u r d e n , as Z e k i p r o r Zeki s confusions _ poses, it w o u l d have to investigate w h y o n e can k n o w h o w b u t c a n n o t believe h o w ; w h y o n e asks ' H o w d o y o u k n o w ? ' and ' W h y d o you believe? 1 , b u t n o t ' W h y d o y o u k n o w ? ' or ' H o w d o y o u believe?'; w h y o n e can k n o w w h e t h e r or better b u t c a n n o t believe w h e t h e r or better; w h y k n o w i n g that p a n d believi n g that p are n o t related in the w a y that k n o w i n g N N a n d believing N N are related; w h y ' H e believes that p, b u t it is false that p' is in order, b u t 'I believe that p, b u t it is false that p' is a k i n d of nonsense; a n d so o n . C o u l d k n o w l e d g e o f t h e functioning of the brain conceivably shed light o n these logico-grammatical questions? Z e k i appears to believe that these problems fall w i t h i n the province of neuroscience because o f ' t h e exquisite capacity of the h u m a n brain . . . to acquire k n o w l e d g e ' . B u t , as s h o u l d b y n o w b e evident, the brain has n o capacity to acquire k n o w l e d g e at all - it is h u m a n beings w h o have that capacity. It is, to b e sure, correct that w e w o u l d n o t have any such capacity b u t for the k i n d o f brain w e have. A n d it is true that n e u r o s c i e n c e has t h e task o f studying t h e brain, and h o p e s to c o m e to understand its structure a n d functioning, w h i c h e n d o w us w i t h this capacity. B u t that is n o reason for supposing that n e u r o s c i e n c e can answer conceptual questions c o n c e r n i n g t h e relationship b e t w e e n k n o w l e d g e a n d belief. It is equally confused to suppose that neuroscience has Nor can neuroscience resolve the ., , , , ., , . , . (1~. , ,. provided answers to such philosophical questions as D o r status of secondary qualities , , , ,,,, , _, . colours exist in the material world? a n d C a n colours be considered to b e t h e properties o f objects?' T h e answers w h i c h Z e k i supposes t o have b e e n given by neuroscience w e r e already given in essence b y Galileo, Descartes, Boyle a n d Locke in the seventeenth century. T h e answers, that Z e k i takes neuroscience to offer are just as questionable as t h e s e v e n t e e n t h - c e n t u r y metaphysics w h i c h anticipated t h e m .

Methodological

Reflections

407

The claim that neuroscience has shown that objects have ^e ' ™ no colours is evidently confused, at least if taken at face value. For, literally understood, this would imply that all objects are colourless — like transparent, colourless window panes. What is, or should be, meant is that it is meaningless to ascribe colours to material objects — that material objects could not be coloured. But how could neuroscience show that something is logically impossible? In particular, how could it show that no extended spatial object could have a colour? The suggestion that colours are properties of the brain is evidently confused, for, of course, Zeki does not mean that the brain is grey to white in colour, and that nothing else has any colour at all. What he must mean is that the fact that most material objects appear to us to have a colour is due to the functioning of our brains. But all that neuroscience can do is discover the mechanisms that enable us to apprehend the colours of things — and that in no way shows things not to be coloured, let alone that it is logically impossible that they have a colour. Zeki's other claim - that colour is 'the interpretation that the brain gives to that physical property of objects (their reflectance), an interpretation that allows it to acquire knowledge rapidly about the property of reflectance' - is equally confused. For the brain can neither interpret, nor acquire knowledge of, anything - a fortiori not of the reflectance properties of things. The person whose brain it is may acquire knowledge of the colours of the things he sees as a result of processes in his brain, but it does not follow that he acquires any knowledge about the reflectance properties of those objects. For, unless he knows some physics, he is unlikely to know anything about reflectance properties and light wavelengths. But this does not show that the colours he takes himself to see are not actually the colours of objects. After all, if that were so — that is, if colours were not objective properties of objects - then he would not be able to see anything in his environment at all, since what has no colour (a fortiori what can have no colour) is not visible. Of course, that colours are not objective properties of objects is exactly what Galileo, Descartes, Boyle and Locke argued. According to their representationalist metaphysics, what we really visually perceive are merely ideas in our minds, from which we draw problematic inferences about the imperceptible world in which we presume we live. But is that what twenty-first-century neuroscience has to contribute to philosophy? Such metaphysical doctrines have been with philosophy for 350 years, and whether or not they are defensible, the one thing that should be clear is that they are not empirical questions that could conceivably be settled by neuroscientific investigations. No neuroscientific discoveries can solve any of the con. 1 1 1 , r 1 -1 ceptual problems that are the proper province 01 phiior , , , ,* ,. . . . . . sophy, any more than the empirical discoveries or physicists can prove mathematical theorems. For the description of any discovery in cognitive neuroscience presupposes the relevant psychological concepts.45 Factual discoveries cannot determine what makes sense. They determine what is true — which presupposes what makes sense. No factual discovery can resolve a J , . ' conceptual question

^ Of course, this does not imply that individual neuroscientists may not acquire expertise in philosophical analysis and contribute to conceptual clarification.

408

On Method

Neuroscience has a multitude of great tasks. It aims to achieve an understanding of the neural conditions that endow us with the distinctive human capacities we possess Its investigations into sensation, perception, memory, affection and volition are already shedding light on these subjects. Achievement in cognitive neuroscience is gradually enlarging our understanding of why we are as we are, why we possess the powers we possess, what determines their empirical limitations, and what goes on in our brains when we exercise them. Neuroscientific advances also hold out the hope that in respect of certain fearful afflictions, hitherto conceived to be beyond our powers to treat, it may after all, be possible to ameliorate the human condition.

14.5 Why it Matters . On the question Jof how it will „ , . affect the next experiment

We can imagine a scientist reading our analytical discus. ., , m TT - I I -in. sions with some bafflement. He might be mildly mtera ' ested in some ot our connective analyses, yet nevertheless puzzled at what seems to be endless logic chopping. 'Does all this really matter?', he might query when he has read our opening discussions. 'After all', he might continue, 'how is this going to affect the next experiment?' We hope that any reader who has followed us thus far will not be tempted to ask this question. For it displays incomprehension. Whether our analytic reflections do or do not affect the next experiment is not our concern. They may or may not - that depends on what experiment is in view, and what the neuroscientist's presuppositions are. It should be obvious, from our foregoing discussions, that, if our arguments are cogent, some experiments might best be abandoned (see, for example, our discussion of voluntary movements in §8.2). Others would need to be redesigned (see, for example, our discussion of mental imagery in §6.3.1). Most may well be unaffected, although the questions addressed might need to be rephrased, and the results might need to be described in quite different ways than hitherto (see, e.g., §14.3). Our concern has not been with the design of the next Our concern is with . , , ., , . ,. , r , , ,. , , . experiment, but rather with the understanding or the last 1 understanding the last experiment . . . experiment. More generally, conceptual investigations contribute primarily to understanding what is known, and to clarity in the formulation of questions concerning what is not known. It would not matter in the least if our reflections have no effect on the next experiment. But they do have considerable effect on the interpretation of the results of previous experiments. And they surely have something to contribute to the asking of questions, to the formulation of questions, and to distinguishing between significant and confused questions. (If we are right, then questions about 'the binding problem', understood as the problem of how the brain forms images, are largely expressions of confusion (see §4.2.3), and much of the debate about mental imagery is misconceived (see §§6.3 and 6.3.1).) . , , Does all this apparent logic chopping, all this detailed Does it matter? If understanding ,. . ¿. , , , . . _ , , discussion ot words and their use, matter: Does neuromatters then it mutters science really need this sort of thing? If the moving spirit behind the neuroscientific endeavour is the desire to understand neural phenomena and

Methodological Reflections

409

their relation to psychological capacities and their exercise, then it matters greatly. For, irrespective of the brilliance of the neuroscientists' experiments and the refinement of their techniques, if there is conceptual confusion about their questions or conceptual error in the descriptions of the results of their investigations, then they will not have understood what they set out to understand. Most contemporary neuroscientists working in the domain of cognitive neuroscience agree that Sir John Eccles's advocacy of a form of dualism (see §2.3) was a mistake - and it is a conceptual confusion that líes at the heart of Eccles's error. "We have tried to demonstrate, by reference to a variety of theories of distinguished contemporary cognitive neuroscientists, that conceptual error, far from being eradicated by a superficial rejection of various forms of Cartesian dualism, is widespread. It affects and infects the cogency of the questions addressed, the character of the experiments devised to answer them, the intelligibility of the descriptions of the results of these experiments, and the coherence of the conclusions derived from them. And this surely matters both to the understanding of what current neuroscientists have achieved, and to the further progress of cognitive neuroscience. It also matters greatly to the educated public. For, irreWhv it matters to the educated . . . . _ r f. spective of whether certain neuroscientists are confused, there is no question but that the forms of description they employ confuse the lay public. Neuroscientists are understandably eager to communicate the knowledge they have attained over the past decades about the functioning of the brain and to share with the educated public some of the excitement they feel about their subject. That is evident from the flood of books written by numerous distinguished members of the profession. But by speaking about the brain thinking and reasoning, about one hemisphere knowing something and not informing the other, about the brain making decisions without the person knowing, about rotating mental images in mental space, and so forth, neuroscientists are fostering a form of mystification and cultivating a neuromythology that are altogether deplorable. For, first, this does anything but engender the understanding on behalf of the lay public that is aimed at. Secondly, the lay public will look to neuroscience for answers to pseudo-questions that it should not ask and that neuroscience cannot answer. Once the public become disillusioned, they will ignore the important genuine questions which neuroscience can both ask and answer. And this surely matters. We have, throughout this book, tried to show that clarity On the need for conceptual clarity . .. . concerning conceptual structures is as important tor cognitive neuroscience as clarity about experimental methods. Neuroscience's great contributions to our understanding of the biological roots of human capacities and their exercise are illuminated, not hindered, by such clarification. For only when the long shadows cast by conceptual confusions are chased away can the achievements of neuroscience be seen aright.

Appendices

Appendix 1

Daniel Dennett

Daniel Dennett has written extensively on cognitive science and philosophy of mind. His first book, Content and Consciousness (1969), laid the foundations for, and determined the direction of, his endeavours for the next thirty years. He has investigated two themes: intentionality and consciousness. In his next book, a volume of essays entitled Brainstorms (1978), he pursued both themes further. But the two subjects were handled separately in his best-known works: The Intentional Stance (1987), a volume of essays on intentionality, and Consciousness Explained (1992), a full-length treatment of consciousness that presupposes and builds on his account of intentionality. A more recent volume, Kinds of Minds (1996) brings both themes together again. Dennett studied philosophy at Harvard, where he was taught by W. V. O. Quine, and at Oxford, where he was taught by Gilbert Ryle. He acknowledges a debt to both his teachers. Quine's influence is indeed evident, but Ryle seems to have influenced him either merely negatively (in respect of his criticism of Cartesianism) or through misinterpretation.1 Dennett avers a deep agreement with both his philosophical mentors over the nature of philosophy. 2 This is surprising, since they have diametrically opposed conceptions of the subject, Ryle believing that philosophy is sui generis, and radically distinct from science, Quine believing that it is continuous with science. Dennett writes that his 'debt to Wittgenstein is large and outstanding'. 3 However, Dennett's accounts of intentionality and consciousness are very far indeed from anything "Wittgenstein would have countenanced.

1 Dennett, like many American philosophers, labours under the illusion that Ryle was a logical behaviourist (see 'Three kinds of intentional psychology', repr. in The Intentional Stance (MIT Press, Cambridge, MA., 1987), p. 45; subsequent references in the text to this paper are flagged 'TK'). Despke a few incautious statements that veer in a behaviourist direction, Ryle was certainly not a logical behaviourist (see G. Ryle, The Concept of Mind (Hutchinson, London, 1949), pp. 327-30, and idem, 'Adverbial verbs and verbs of thinking', repr. in his volume of essays On Tliinking (Blackwell, Oxford, 1979), p. 17). Dennett, 'Dennett, Daniel C.\ in S. Guttenplan (ed.), A Companion to the Philosophy of Mind (Blackwell, Oxford, Í994), p. 237. D. Dennett, Consciousness Explained (Penguin, Harmondsworth, 1993), p. 463. Subsequent references in the text to this volume will be flagged 'GE'.

414

Appendix 1: Daniel Dennett

Like Quine, Dennett thinks that 'philosophy is allied with, and indeed continuous with, the physical sciences'.4 Hence he deems himself a 'philosophical naturalist', taking philosophical problems to be soluble 'through a combination of scientific enquiry and the adjustment of our conceptual prejudices in light of empirical evidence'.5 Accordingly, his aim in his investigations of intentionality and consciousness is 'to create defend, and confirm (or disconfirm) theories that are directly about the phenomena'. 6 As a graduate student at Oxford, Dennett decided that he 'had to figure out how the brain could possibly accomplish the mind's work . . . how the mechanical responses of "stupid" neurons could be knit into a fabric of activity that actually discriminated meanings'. He took his task to be 'sketching the outlines of a physical structure that could be seen to accomplish the puzzling legerdemain of the mind'. 7 These are great ambitions - indeed, ambitions that are arguably beyond the province of philosophy and the competence of philosophers. We shall see to what extent Dennett deems himself to have achieved them. Dennett's books are widely read both by educated laymen and by cognitive neuroscientists eager to discover what light philosophers can shed upon their problems. It has been written of him that he 'has played a central role in changing the way we understand the nature of philosophical problems, the nature of philosophy, and the relationship between philosophy and natural science'. His ability to integrate philosophy, psychology and the study of artificial intelligence has, it is said, 'set the tone and direction for contemporary philosophy of mind'. Even those who disagree with him, it is claimed, 'now follow him in acknowledging the role of empirical research in the solution of traditional philosophical problems'. 8 Precisely because we disagree with so much of what Dennett has written, and because he is so widely read by neuroscientists, it is important that we make clear why we disagree with him on general methodology, on the account he gives of what he calls 'the intentional stance', and on his 'theory of consciousness'. It should be evident that, like Ryle and "Wittgenstein, we do not think that empirical research can solve any philosophical problems, any more than it can solve problems in mathematics. So we part company with Dennett right at the beginning, inasmuch as we repudiate the Quinean naturalism that he adopts. W e do not think that any philosophical problems can be solved through scientific enquiry. We acknowledge that concepts sometimes need to be revised or augmented for scientific purposes, but deny that any philosophical problems can be solved thereby, indeed, the suggestion that they can seems to us to be akin to claiming that knots can be untied in a piece of string by taking a fresh piece of string.

4

Dennett, 'Setting off on the right foot', in Intentional Stance, p. 5. The observation is John Synions's, in his book On Dennett (Wadsworth, Belmont, CA, 2002), p. 12. Since the book appears to have Dennett's-imprimatur, we presume he concurs. 6 Dennett, quoted in ibid., p. 15. 7 Dennett, 'Dennett, Daniel C.\ p. 236. 8 Symons, On Dennett, p. 8. 5

Appendix 1: Daniel Dennett.

415

1 Dennett's Methodology and Presuppositions Like the Churchlands, Dennett conceives of our common or garden psychological terms as the vocabulary of'folk psychology', a theoretical edifice that is part of the intellectual legacy of Everyman. Unlike the Churchlands, however, Dennett holds folk psychology to be a true theory, and therefore to be a candidate for incorporation into 'science'. (TK 47f). For, in his view, folk psychology works: it is 'thanks to folk psychology [that] we cooperate on multi-person projects, learn from each other and enjoy periods of local peace'. 9 Nevertheless, he argues, before we can incorporate such a proto-theory into a genuine science, we must separate the wheat from the chaff. We must discern what it is within the theory that accounts for its explanatory and predictive success. 'In this way we can criticize as we analyse, and it is even open to us in the end to discard folk psychology if it turns out to be a bad theory, and with it the presumed theoretical entities named therein' (TK 47). Dennett holds that folk psychology 'can best be viewed as a sort of logical behaviourism': what it means to say that someone believes something is that he is disposed to behave in certain ways under certain conditions, namely, in ways in which it would be rational for him to behave given his other beliefs and desires (TK 50). However, he also claims that it is part of the theory of folk psychology that beliefs 'are information-bearing states of people that arise from perceptions and that, together with appropriately related desires, lead to intelligent action' (TK 46). He further holds that folk psychology is a 'rationalistic calculus of interpretation and prediction — an idealizing, abstract, instrumentalistic interpretation method that has evolved because it works' (TK 48). It is far from evident how the concept of belief can simultaneously signify or be thought to signify an informationbearing state, a mere behavioural disposition, and also no more than an instrumentalist predictive device. "We shall revert to this point below. As noted, Dennett aspired to contribute to a substantial, first-order, empirical theory about human psychology. So, 'from the outset', he avers, 'I worked from the third-person point of view of science'. 10 This methodological commitment was derived from the behaviourists, according to whom 'only facts garnered "from outside" count as data. . . . The idea at its simplest was that since you can never "see directly" into other people's minds, but have to take their word for it, any such facts as there are about mental events are not among the data of science, since they can never be properly verified by objective methods. This methodological scruple . . . is the ruling principle of all experimental psychology and neuroscience today (not just "behaviourist research")' {CE 70). Nevertheless, Dennett points out, even if mental events are not among the data of science, this does not mean that we cannot study them scientifically. The challenge, as he sees things, is to construct a theory of mental events, using the data that scientific method permits. 'Such a theory will have to be constructed from the third-person point of view, since all science is constructed from that perspective' (CE 70).

9 1(1

Dennett, 'Setting off on the right foot', p. 11. Dennett, 'Dennett, Daniel C.\ p. 237.

416

Appendix Í: Daniel Dennett

Dennett has no qualms about ascribing psychological properties to the brain. He thinks that the brain is conscious (CE 172), that it gathers information from the world and uses it to extract anticipations (CE 144). The brain, he avers, often makes simplifying assumptions, makes use of supporting information, arrives at conclusions, and interprets the information it receives (CE 142f.). This, as we have argued (see chapter 3), is incoherent. Dennett also claims that the brain is turned into a mind as a result of being parasitized by what he calls 'memes' (CE 254). We shall investigate this claim below. Given that he believes that the brain becomes and then is the mind, it is surprising to find him also claiming that each of us is more intimately acquainted with our mind than with our brain. For if the mind is the brain, then to be acquainted with the mind is to be acquainted with the brain (even if one does not realize it). Dennett goes further: so intimately are we acquainted with our minds, he avers, 'that you might even say that you are your mind'. That thought suggests, in his view, that 'each of us knows exacdy one mind from the inside, and no two of us know the same mind from the inside'. 11 It is beyond dispute, he claims, that we have minds (although it remains unclear how we can both be and also have a mind). Like Descartes, Dennett holds that we know that other people have minds because they can understand speech, and only things with minds can do that (KM 8). Wherever there is a mind, he claims, 'there is a point of view. This is one of the most fundamental ideas we have about minds - or about consciousness. . . . For most practical purposes, we can consider the point of view of a particular conscious subject to be just that: a point moving through space-time' (CE 101, our emphasis). Before moving on to the substantive parts of Dennett's philosophy, we should pause in order to clarify our disagreements on two large methodological issues: folk psychology and points of view. Folk psychology As we have argued, our ordinary psychological vocabulary is not a theoretical vocabulary, and it is not part of a rudimentary proto-theory (§§13.2.1-13.2.2). Our ordinary use of words such as 'see', 'hear', 'know', 'believe', 'like', 'dislike', 'pain', 'pleasure', 'anger', joy' does not disclose a theory of psychology — it exhibits the concepts that characterize sentient creatures, in particular human beings. These concepts are not theoretical ones. The nomináis 'perception', 'knowledge', 'belief, 'attitude', 'emotion, 'liking', 'pleasure', 'anger' are not names oí theoretical entities, since they are neither names of entities nor theoretical. The idea that we could, in principle, discard this vocabulary (if 'folk psychology' turned out to be a 'bad theory') is, as we have argued, confused (§§13.2.3-13.2.4). 'Logical behaviourism' is a philosophical doctrine according to which sentences containing psychological terms are logically equivalent to, and translatable into, sentences about behaviour and dispositions to behave. It is, therefore, incoherent to characterize socalled folk psychology - that is, the vocabulary of the mental and the forms of explanation of action and emotion that invoke it - as a form of logical behaviourism. At most it might

" D. Dennett, Kinds of Minds, (Weideufeld and Nicolson, London, 1996), p. 3. Subsequent references in the text to this book will be flagged LKM\

Appendix 1: Daniel Dennett

All

be claimed that logical behaviourism gives a correct analysis of ordinary psychological predicates. That would be false. {It makes perfectly good sense to describe someone as thinking something but not saying or being disposed to say what they think, or to describe someone as saying something without believing what they say.) But, of course, it is not a claim made by so-called folk psychologists — that is, by the normal competent users of our common psychological vocabulary. Dennett's suggestion that folk psychology conceives of beliefs as information-bearing states of people is misguided. In the first place, the logical grammar of the verb 'to believe' is not that of a verb signifying a state. Secondly, and consequendy, ordinary ascriptions of belief and explanations of action by reference to beliefs do not imply that the person to whom beliefs are ascribed is in any kind of mental state. Our common or garden belief ascriptions are not mental state ascriptions (see §10.2). Dennett's suggestion that folk psychology is 'a rationalistic calculus of interpretation and prediction — an idealizing, abstract, instrumentalistic interpretation method' is misconceived. When we teach children to prefix T think' or 'I believe' to sentences or sentence nominalizations (i.e. that-clauses) or to insert these phrases parenthetically within a sentence ('Daddy is, I believe, in town'), we are not teaching them an 'abstract instrumentalistic interpretation method'. We are teaching them, for example, to indicate the character, or degree of adequacy, of their grounds for assertion. When we report someone's assertion, and prefix or parenthetically interpolate 'she believes' or 'he thinks', we are not interpreting or predicting anything, but typically distancing ourselves from what that person is reported as having asserted ('She believes that N N is plotting — but (I know) he is not'). When we teach a child to say 'It hurts' and later 'I have a pain', we are not teaching it part of an instrumentalist predictive calculus, and when the child learns to ascribe pains to his mother, and to commiserate with her when she has hurt herself, he is not learning how to operate an 'instrumentalistic interpretation method'. And when we teach a child to replace his cries of fear with 'Mummy, I had a nightmare!', we have not taught him a method of interpretation - he can learn that when he studies psychoanalysis. Finally, it is misconceived to think, as Dennett does, that 'it is thanks to folk psychology [that] we cooperate on multi-person projects, learn from each other, and enjoy periods of local peace'. We do not cooperate with each other because we share a theory about 'other minds', any more than members of a wolf pack cooperate in hunting because they share a theory about 'other [wolf] minds'. Babies do not learn from their parents because they have worked out a folk-psychological theory about their parents. And we do not enjoy periods of local peace, any more than we suffer periods of local strife, because we have mastered the vocabulary of the mental. There is no such thing as a theory of'folk psychology', save for such dicta as 'Once bitten, twice shy' or 'Spare the rod and spoil the child' (see §§13.2-13.2.2). It is not that folk psychology 'works' - it is no theory that might work or fail to work, and such commonplace dicta are as often as not misguided. Rather, the vocabulary of the mental pardy defines what it is to be human - but that is an altogether different matter. Of course, knowing what another person knows and believes, knowing what he wants or intends, and knowing his emotions and attitudes, enables modest degrees of prediction — but that is not because of any theory of human behaviour. It is because of the limited rationality we enjoy and the character of the practical reasoning we engage in.

418

Appendix í: Daniel Dennett

Points oj view The expressions 'point of view', 'first-person point of view', and 'thirdperson point of view' are 'ready-mades' of current cognitive science and philosophy of mind. The thought that one of our most fundamental ideas about the mind is that wherever there is a mind there is a point of view dresses up distorted trivialities as profundities. At best all it amounts to is (a) that sentient creatures perceive whatever they perceive from a particular place, (b) that different people may have different opinions about certain things, and (c) that language-using creatures can give expression to their thoughts and experiences using the first-person pronoun. The last triviality does not signify that first-person expressions, or even reports, of thought and experience are made from 'a point of view'. H have a headache' or 'I think that Shakespeare died in 1616' is no more a statement made from 'a point of view' than is '1 am six foot tall'. Can one intelligibly say 'From my point of view, Í have a headache', or 'From my point of view, I think Shakespeare died in 1616'?

The claim that 'science' works 'from a third-person point of view' is explained by reference to behaviourism. On that account, 'only facts garnered "from outside" count as data'. The metaphor is unfortunate. Outside what? The mind is no place, and the apparently locative phrases 'in one's mind', 'at the back of one's mind', 'crossed one's mind' are misleading forms of words that are readily unpacked into verb phrases that make no use of the nominal 'mind' (see §3.10) or of the apparently spatial expressions 'in', 'at the back of, and 'crossed'. To say that a thought crossed one's mind — that is, that one thought of something - or that something is fixed indelibly in one's mind - that is, that one will never forget something — is not to report 'from inside' anything, and to assert third-person sentences such as 'He thought of something' or 'He never forgot the searing experience' is not to report anything 'from outside'. The metaphor of 'inside' and 'outside' wreaks havoc with our thought if we misinterpret it (see §§3.5-3.7). The putative contrast between the third-person point of view of'science' and the firstperson point of view is explained by reference to the claim that one can never see directly into another person's mind, but has to take their word for it, and therefore one can never properly verify what people say about their thought and experience. According to Dennett, this methodological scruple informs all contemporary experimental psychology and neuroscience. Such scruples are misconceived. It is wrong to think that one cannot properly verify whether another person knows or believes something - we do so all the time. (What, to take but one example, are examinations for?) It is mistaken to suppose that one cannot verify whether another is in severe pain - just try doubting the screams of a person badly injured in an accident. There is no difficulty at all in verifying that little Tommy likes ice-cream. Psychiatrists successfully verify whether someone has a depression, and opticians can readily check whether someone sees double. And much of what neuroscientists have discovered about functional localization in the brain involved the data provided by patients' avowals during brain operations (and., more recently, brain scans). The sense in which one cannot 'see directly' into another's mind is the sense in which one cannot see sounds or hear colours: namely, because there is no such thing — this is a senseless form of words (hence it cannot describe a condition for knowing what another person thinks or feels). Equally, of course, there is no such thing as 'seeing directly' into one's own mind - that notion incorporates a confused conception of introspection (see §3.6). A

Appendix 1: Daniel Dennett

419

person's utterances 'I can hear such-and-such', 'I see such-and-such', 'I think it is thus', 'I fear that so-and-so', 'I wish you would go away', l I ani tired', etc. are not reports derived from his observations of a peep-show which only he can see. He cannot hear his hearing or see his seeing, although others can often observe that he sees or hears something and. indeed, see or hear what he sees and hears. Finally, if it were so difficult, or indeed impossible, to see, detect or find out another person's thoughts and feelings, then it would be a mystery that from time to rime we have to make such an effort to conceal what we think and feel from our audience. First-person psychological utterances may be expressions of thought or experience, or may be reports or descriptions. But they are not based on perceiving something that no one else can see, or on having 'access' to something to which others have no 'access'. They are respectable data for medical or psychological diagnosis, and for neurosci en tifie investigation. That is perfectly compatible with the thought that self-deception is possible, that people sometimes lie or are confused — and that we are accordingly sometimes deceived or mistaken in our judgements about them. It is quite wrong that we have to take others' word for what they know or believe, feel or intend — we can often detect confusion, deception and self-deception. 'Science' does not investigate a subject matter from either a first- or a third-person point of view. But one might say that meteorologists and climatologists investigate the melting of the polar ice-caps from a physical and climatological point of view, whereas zoologists investigate it from a zoological point of view, and ecologists study it from an ecological point of view. Doctors may investigate the so-called madness of George III from a medical point of view (the poor man, as has been discovered, suffered from porphyria); historians may study it from a political-historical or from a medical-historical point of view. There is no single, uniform subject called 'science' that is pursued by all scientists in their investigations, any more than there is a single subject matter called 'reality' that furnishes them with a subject matter. There are numerous sciences, each with different concerns. Scientists, at their best, are dedicated to the pursuit of truth within their subject. Their data are the facts that they establish within their domains. These are selected according to their scientific points of view - the relevant data for a physicist being very different from the data relevant to the interests or concerns of a palaeontologist, ecologist, zoologist or climatologist. In the psychological sciences, the evidential support required to establish data includes the subjects' avowals and averrals of thought and experience. These are not mere words — nor are they reports of a private peep-show.

2 The Intentional Stance The expression 'the intentional stance' is introduced by Dennett to characterize a 'tactic' 12 or a 'strategy' (KM 27) of interpretation. H e contrasts it with what he calls 'the physical

'2 Dennett, 'Dennett, Daniel C , p. 239.

420

Appendix Í: Daniel Dennett

stance' and 'the design stance', the former being exemplified by physics, the latter by our descriptions of artefacts. It is a mode of interpreting 'entities' (people, animals, artefacts) as if they were rational agents with beliefs, desires and other mental states exhibiting what Brentano called 'intentionality'. 13 Dennett explains this term of art by equating it with what he calls 'aboutness'. Something is held to exhibit intentionality 'if its competence is in some way about something else', in Dennett's view, 'A lock and key exhibit the crudest form of intentionality; and so do the opioid receptors in brain cells' (KM 35). Perceptual and emotional states, as well as 'states of memory' are intentional. A 'state of recognition' 'exhibits a very particular aboutness', inasmuch as one recognizes whatever one recognizes as something — for example, a horse as a horse. Had one recognized it as something else - for example, a moose or a motor cyclist - one's perceptual state would have had a different aboutness (KM 36f.). The intentional stance is, he claims, the attitude or perspective we routinely adopt whenever we attribute intentional mental states to ourselves and to others. According to Dennett, the fundamental form of attributions of mental states 'are sentences that express what are called prepositional attitudes'. Such sentences have the general form 'A V's that p\ where 'V holds a place for a verb signifying a 'propositional attitude', and the variable 'p' is 'a term for a particular content or meaning of that attitude — the proposition denoted (e.g. 'Jack believes that it will rain'). Propositions, in Dennett's view, 'are the theoretical entities with which we identify and measure beliefs. For two people to share a belief is, by definition, for them to believe one and the same proposition' (KM 45f.). This is a misconception. The sentences 'A believes that it will rain', ! B fears that it will rain', 'C hopes that it will rain , and 'D suspects that it will rain' do not signify attitudes to propositions. It is, of course, possible to believe propositions, just as one can believe statements, stories and rumours. But since A may believe what B fears, and also what C hopes and D suspects, and since there is no such thing as fearing, hoping or suspecting a proposition, what A believes (in our example) is not the proposition that it will rain, but simply that it will rain. "We do not measure beliefs, fears, hopes and suspicions by propositions, but we individuate them by specifying what is believed, feared, hoped or suspected — which cannot be a proposition in the latter three cases, and need not be one in the case of belief. Finally, propositions are not theoretical entities, any more than are statements, declarations, stories, fairy-tales or rumours. The intentional stance, according to Dennett, is an interpretative stance.14 We interpret an entity thus by adopting the presupposition that it approximates the ideal of an optimally designed (i.e. rational} self-regarding agent, governing its choices by reference to its beliefs and desires. We do so, Dennett claims, in order to predict and explain its actions or moves. Adopting the intentional stance toward something other than a human being, he

13

Dennett,'True believers: the intentional strategy and why it works', repr. in Intentional Stance, p. 15. Subsequent references in the text to this article are flagged *TB\ 14 In this respect it seems to stand in contrast with what Dennett calls 'the physical stance'. "When physicists describe and predict natural phenomena, they are not interpreting them as if they were correctly described by the laws of physics. Physics is not an interpretative stance.

Appendix 1: Daniel Dennett

421

observes, 'seems to be deliberately anthropomorphizing it' {KM 21). Nevertheless, he thinks, it is wholly warranted if that thing is an intentional system. 'Intentional systems' are entities whose behaviour is predictable and explicable from an intentional stance. Strikingly, Dennett thinks that such heterogeneous items as selfreplicating macromolecules, thermostats, amoebas, plants, animals, people and chessplaying computers are intentional systems (KM 34). The strategy is to think of these things as rational agents: 'you figure out what beliefs that agent ought to have, given its place in the world and its purpose. Then you figure out what desires it ought to have on the same considerations, and finally, you predict that this rational agent will act to further its goals in the light of its beliefs' (TB 17). This strategy, according to Dennett, works with people almost all the time. . . . The strategy works on birds, and on fish, and on reptiles, and on insects and spiders, and even on such lowly and unenterprising creatures as clams. . . . It also works on some artefacts: the chess-playing computer wiil not take your knight because it knows that there is a Une of ensuing play that would lead to its losing its rook, and it does not want that to happen. More modestly, the thermostat will turn off the boiler as soon as it comes to believe the room has reached the desired temperature. (TB 22) Why does the strategy work? In the case of human beings, Dennett contends, it works because 'evolution has designed human beings to be rational, to believe what they ought to believe and to want what they ought to want' (TB 33). Nevertheless, attributions of what Dennett calls 'intentionality' are no more than 'interpretations of the phenomena — a "heuristic overlay", describing an inescapably idealized pattern. Like such abstracta as centers of gravity and parallelograms offeree, the beliefs and desires posited by the highest stance have no independent and concrete existence.' 15 But for all that, the intentional stance is held to be unavoidable, both with regard to oneself and with regard to one's fellow intelligent beings (TB 27). There are multiple objections to these views. We shall first examine Dennett's conception of intentionality and then his idea of the 'intentional stance'. Intentionality Dennett misconstrues what modern philosophers since Brentano (who revived the medieval term 'intentio') have called 'intentionality'. Brentano held (wrongly) that intentionality is the mark of the mental. Mental attributes, he argued, have 'objects'. If a person V's (e.g. believes, hopes, fears, suspects, expects, loves, hates), then there is something that he Vs. What he V's need not exist, occur or be the case in order for it to be true that he V's it. According to Brentano, mental phenomena are unique in respect of 'the intentional in-existence of [their] object' - one can believe, fear or suspect that something or other is going to happen, but it may never come to pass. But the non-occurrence of what one believed, feared or suspected would occur does not imply that one believed, feared or suspected nothing. The object of one's belief, fear

Dennett, 'Dennett, Daniel C , p. 239.

422

Appendix

1: Daniel

Dennett

or suspicion is an 'intentional object'. S o m e w h a t differently, o n e m a y l o o k for (but n o t find) E l d o r a d o even t h o u g h it does n o t exist, or fear (but n o t m e e t ) ghosts even though there are n o n e . M e n t a l p h e n o m e n a , B r e n t a n o held, 'contain an object intentionally within themselves', w h i c h is said to enjoy immanent existence w i t h i n , for example, the belief, fear or suspicion. 1 6 N o d o u b t this was confused. First, there are psychological p h e n o m e n a that are not intentional at all, such as objectless m o o d s (e.g. cheerfulness) a n d sensations (e.g. pain). Secondly, it is arguably mistaken to suppose that t h e r e is any f o r m of intentional i n existence a b o u t believing Jack or his story (by contrast w i t h fearing G o d (or Zeus) and his wrath). For in so far as Jack must exist in order for o n e to believe him, a n d in so far as he m u s t have told a story i n order for o n e to believe it, these are n o t intentional objects of belief. T h e y m a y b e contrasted w i t h believing that Jack is guilty o r suspecting treachery (i.e. suspecting that there is treachery), the intentional objects o f w h i c h are specified by nominalization accusatives (i.e. that-clauses or their equivalents), as well as fearing Zeus or looking for Eldorado. So n o t all forms of m e n t a l attributes (such as belief) that can intelligibly b e said to b e intentional are intentional in all occurrences. Thirdly, it was misleading to speak of the 'intentional in-existence' of intentional objects, w h e n all that was necessary was to p o i n t o u t that for certain psychological attributes, o n e can Vthat p even t h o u g h it is n o t the case that p, and o n e can V M even t h o u g h there is n o M . W e need n o t delve further i n t o the logical complexities here, 1 7 for this r o u g h indication suffices to show deficiencies in D e n n e t t ' s discussion of the matter. It is u n i l m m i n a t i n g t o characterize intentionality as 'aboutness': if o n e fears defeat and h o p e s for victory, one's fear is n o t about defeat, a l t h o u g h its intentional object is that one will be defeated, and one's h o p e is n o t about victory, a l t h o u g h its intentional object is that one will be victorious. T h e preposition 'about' a n d the barbaric n o m i n a l 'aboutness' are ill-suited to capture w h a t B r e n t a n o a n d the scholastics m e a n t b y 'intentionality' or to p i n p o i n t the intentional objects w h i c h they c o n c e i v e d o f as 'existing i m m a n e n t l y ' in mental p h e n o m e n a . F o r o n e cannot say ' M y belief is a b o u t that it will rain tonight', and the answer to the question ' W h a t d o y o u fear? is n o t ' A b o u t defeat', b u t 'Defeat' or ' T h a t w e shall b e defeated'. It is incorrect to ascribe intentionality to h u m a n beings as such - they d o n o t enjoy any 'aboutness' or 'contain objects intentionally w i t h i n themselves'. T o speak of h u m a n beings

16

F. Brentano, Psychology from an Empirical Standpoint (Routledge, London, 1995; first published 1874), pp. 88f. What Brentano seems to have meant was as follows. W h e n one, for example, believes falsely that p, then what one believes (viz. that p) does not obtain or exist. But it does not follow, of course, that one believes nothing. Moreover, one can say what one believes, even though what one believes does not exist 'in reality'. But then h o w can one say (and h o w can one know) what one believes? According to Brentano we can know, by means of 'inner perception', that we believe, and w e can 'read off' what we believe from our belief. So our belief contains its object 'intentionally' within itself; or, again, the object of our belief enjoys 'immanent existence' within our belief. 17 For an overview of the problems of intentionality, see P. M. S. Hacker, 'An orrery of intentionality', Language and Communication, 21 (2001), pp. 119-41.

Appendix 1: Daniel Dennett

423

'enjoying intentionality' 18 can, at best, only mean that human beings have intentional attributes: for example, that they believe, hope, fear, suspect, etc. that something or other is or will be the case. What is intentional, if anything, is the psychological attribute that has an intentional object (e.g. believing that p, as opposed to believing Harry or his story). One cannot intelligibly ascribe 'intentionality' to molecules, cells, parts of the brain, thermostats or computers. Not only is it a subclass of psychological attributes that are the appropriate bearers of intentionality and not animals or things, but, further, only animals, and fairly sophisticated animals at that, and not parts of animals, let alone molecules, thermostats or computers, are the subjects of such attributes. For, as we have argued at length (see chapter 3), it makes no sense to ascribe belief, fear, hope, suspicion, etc. to molecules, cells, the brain or its parts, thermostats or computers. There is no such tiling as a thermostat believing that it is too hot, thinking that it is cold, wondering whether it will cool down, being convinced that it is hot enough, deciding that the temperature has fallen enough, judging or misjudging. Such an artefact cannot (logically cannot) satisfy the criteria for belief, thought, wonder, conviction, deciding, judging or misjudging. It makes no literal sense to ascribe such psychological attributes to artefacts — 'The thermostat thinks that it is too hot' is a form of words which has no use (other than as a joke). Thermostats can no more think than V-l can get married. A very rich array of possible forms of behaviour in the circumstances of life is already presupposed before a creature can intelligibly (truly or falsely) be said to think. It is equally misconceived to attribute intentionality (or 'aboutness') to locks and keys or to opioid receptors. The fact that a key fits a lock does not imply that there is any sense in which the key is 'about' the lock, let alone that the key has intentional attributes (such as the belief, hope or fear that it will or will not fit a lock) that have intentional objects possessing 'intentional in-existence'. Opioid receptors are no more about opioids than cats are about dogs or ducks are about drakes. A different mistake (see p. 420) is to think that an attribute such as recognition is intentional. If jack recognizes Jill going up the hill, there is nothing intentional about his recognition. His recognition is not about Jill (or anything else) - it is ofJill. But he could recognize Jill only if she exists and is there to be recognized — so Jill cannot be said to enjoy intentional in-existence or immanent existence within jack's recognition. Similarly, he could recognize that Jill is there only if she is actually there; if it is false that she is there, then Jack did not recognize Jill but merely thought he did. In short, recognition is not an intentional attribute. O f course, Jack may recognize Jill, but not recognize her 'as Jill'. He may recognize her as the girl who used to go up the hill, but he may no longer remember her name. But this feature should not be confused with intentionality. They are connected, but distinct. And each has to be disentangled separately. 18

Dennett writes that 'all the intentionality we enjoy is derived from the more fundamental intentionality of these billions of crude intentional systems [presumably macromolecules]' [KM 55). We do not 'enjoy intentionality', although we do believe that things are thus-and-so (even if they are not), fear or suspect that things will be so (even if our fears and suspicions are never realized). The intentionality of our beliefs is not 'derived' from anything, although we doubtless would have no beliefs at all if our brains were not functioning appropriately.

424

Appendix 1: Daniel Dennett

The intentional stance The notion of an intentional stance is apt to characterize various forms of animism distinctive of pre-scientific thought. But to ascribe pain, liking, disliking, perceiving, misperceíving, anger, fear, joy, knowledge, belief, memory, imagination, desire, intention, and so on, to living beings, in particular to human beings, is not to adopt an interpretative stance. Dennett characterizes the intentional stance as a mode of interpreting entities as if they were rational agents with beliefs, desires and other intentional mental states. It is, he holds, a 'heuristic overlay' of a belief/desire calculus that is relative to given goals, and that assumes self-regarding rationality. It is a theoretical posit with neither more nor less reality than centres of gravity or parallelograms of force. Its warrant is its instrumentalist, predictive and explanatory success. But, (i) anthropomorphizing pet owners apart, we do not treat animals as if they were rational agents — since we know perfectly well that they are not. But we do ascribe a wide range of perceptual, affective, cognitive and volitional attributes to animals in a perfectly literal sense. Being a rational agent is not a precondition for the applicability of psychological attributes to a creature. (ii) We do not treat the higher animals as if they were, from occasion to occasion, angry, frightened, contented, or as if they could see and hear, feel and smell. For we know that eagles can see farther than, we can, that dogs have a better sense of smell than we do, and that short-eared owls can hear far better than we can. We know that the fleeing fox is afraid of the hounds, that the snarling dog is angry, and that the purring cat is contented. There is no as if about it. (iii) Dennett claims that the intentional stance is unavoidable, 'with regard to [both] oneself and one's fellow intelligent beings' (TB 27). But to ascribe psychological attributes to oneself, to avow that one is in pain ('It hurts!', 'I have a terrible toothache!'), sincerely to express one's regret or remorse ('I am so very sorry!'), to manifest one's pleasure and joy ('I am so pleased!', 'I am overjoyed'), to voice one's beliefs, fears, hopes, suspicions ('I don't believe a word he said', 'I fear disaster', 'I hope it works', 'I suspect treachery'), to declare one's wants, likings, intentions and so forth, is not to adopt an interpretative stance of any kind towards oneself- it is to give expression to one's pain, regret, pleasure, hopes and fears. And the pain or joy one feels, the beliefs one passionately embraces, the fears and suspicions one harbours, are not heuristic overlays, and are not theoretical posits of any kind. One may agree with Dennett that 'the beliefs and desires posited by the highest stance [i.e. the intentional stance] have no independent and concrete existence', if all one means thereby is that there are no beliefs without believers, no desires without desirers, and that belief and desire are not tangible objects. But it is evident that that is not what Dennett means when he writes of positing beliefs and desires, and compares the psychological attributes to centres of gravity or parallelograms of forces. (iv) The intentional strategy requires one to-think of the subject as a rational agent, and to 'figure out what beliefs that agent ought to have, given its place in the world and its purpose'. But if so, the strategy could not possibly be applied to such things as molecules, brains and their parts, thermostats or chess-playing computers. For it makes no sense

Appendix i: Daniel Dennett

425

whatsoever to suppose that self-replicating molecules ought to believe anything, any more than it makes sense to suppose that negative numbers ought to believe that when multiplied by themselves they yield positive numbers. Given the place of thermostats in the world and their purposes, it is clear that there could be no such thing as thermostats believing anything, hence too, no such thing as something that they ought to believe. Of course, we tend to anthropomorphize our chess-playing computer. We might be inclined to say that the computer 'will not take your knight because it knows that there is a line of ensuing play that would lead to its losing its rook, and it does not want that to happen' (TB 22). But what does this amount to? It is no more than afagon de parler. We know that the computer has been designed to make moves that will (probably) lead to the defeat of whomever plays with it - and there is no such thing as the computer's knowing or wanting anything. And in order to predict its moves, we need not absurdly ascribe knowledge or wants to it, but need only understand the goals of its program and programmer (viz. to make a (mindless) chess-playing machine). For design is one form of teleology, and teleology is a basis for prediction. (v) It is surely misconceived to suppose that in 'adopting the intentional stance' - that is, in ascribing psychological attributes to other human beings - we are supposing that 'evolution has designed human beings to be rational, to believe what they ought to believe and to want what they ought to want', that in so doing we presuppose that we are 'optimally designed (i.e. rational) self-regarding agents' and that that is why our 'intentional strategy' works. Evolution has not designed anything - Darwin's achievement was to displace explanation in terms of design by evolutionary explanations. More importantly, we do not, in our efforts to describe, understand and sometimes also to predict the behaviour and reactions of others, assume that they believe what they ought to believe and want what they ought to want. Alas, we know perfectly well that they often believe and want what they have been taught, told, brainwashed and cajoled into believing and wanting by parents, teachers, priests, gurus, governments and advertising agencies, much of which ought not to be believed or wanted. In ascribing psychological attributes to human beings and in predicting or explaining their behaviour, we do not presuppose that they are uniformly self-regarding creatures. Our conception of rationality is not a self-regarding one. It is not irrational to sacrifice oneself or one's interests for the sake of a greater good, and deeds of quotidian generosity, common modest forms of selflessness, and heroic self-sacrifice are not unpredictable or unintelligible forms of deviance for not fitting into Dennett's market-oriented conception of rationality. Even in the third person, our employment of the psychological vocabulary and the intentional idiom is not guided exclusively, or even largely, by a constant desire to predict the behaviour of others. Curious, interested observation, as well as sympathetic and empathetic participation in the affairs of others play at least as great a role; and that often involves a need to know about the beliefs and desires of others (and if we ask them, they often reveal their beliefs and wants to us), not in order to predict their behaviour, but to advise them on what would be for the best, or simply to understand what they have done. And since we are eyes and ears to each other, we commonly share our beliefs and fears,

426

Appendix i: Daniel Dennett

suspicions and hopes, or engage in painstaking discussion, not in order to predict each other's behaviour, but to discover the truth, or to share our knowledge or our prejudices, or to foster a common sense of identity and community. (vi) Dennett characterized the intentional stance as interpreting the behaviour of an entity (person, animal, part of an animal, or artefact) by treating it as if it were a rational agent with an array of beliefs and desires. In the case of things other than us, this involves anthropomorphizing the thing. But we are also supposed to interpret each other as if we had beliefs and desires - and the beliefs and desires that we thus ascribe to each other are mere theoretical posits {akin to centres of gravity). But this generates an incoherence. In order to be able to treat a being as if it believed that p or wanted to V, we must know what it is to believe or to want something. We must know under what circumstances it is appropriate to judge that another person actually believes or wants something, and indeed when it is appropriate to avow a belief or desire in our own case. Only when we have grasped this - that is, when we have understood the use of the verbs 'to believe' and 'to want' (in both the first and the third person) — can we come to understand what it is to treat a being as if it believed this or wanted that. But on Dennett's account the adoption of the intentional strategy towards an intentional system, including human beings, is never other than an 'as if. But if so, there is no actuality in terms of which we might intelligibly cash the pretence, no believing or wanting in terms of which to understand what we are to feign when we treat a being (including ourselves and others) as if they believed or wanted things. Moreover, there is no specification of what would count as actually believing or wanting. But if so, we can assign no content to the concepts of belief and desire that Dennett deploys. (vii) Dennett attempts to explain intentionality as an interpretative strategy. But 'to interpret something as. . .' is itself an intentional expression. To interpret something as something is to take it to be so, to believe or suppose it to be so. But now the 'intentional stance' patently involves an incoherence, for it cannot be applied reflexively. Given Dennett's conception of the intentional stance, it is unclear what precisely he means by claiming that the brain gathers information, anticipates things, interprets the information it receives, arrives at conclusions, etc. Presumably he is 'adopting the intentional stance' towards the brain, and is treating it as if it were a rational agent that believes what it ought to believe and desires what it ought to desire and acts on its beliefs and desires. But this is not coherent. We know what it is to treat a young child as if it were an adult, rational human being, but do we have any idea what it would be to treat a brain as if it were a rational being? The brain, as we have argued, is not a possible subject of beliefs and desires; there is no such thing as a brain acting on beliefs and desires, and there is nothing that the brain does that can be predicted on the basis of its beliefs and desires. Similar obscurity surrounds Dennett's remark that we are intimately acquainted with our own minds, that we know that we have a-mind, and that we know our own mind from the inside. Does this mean that we should, for the sake of predictive efficiency, treat ourselves as if we had a mind, and as if we knew our own mind? To treat a creature as if it had a mind, one must know how to treat a creature that does have a mind. Moreover,

Appendix 1: Daniel Dennett

427

if the mind is merely an instrumentalist calculating convenience, like the centre of gravity in mechanics, then there can be no such thing, even figuratively, as 'knowing, from the inside, that we have a mind'.

3 Heterophenomenological Method It is against the backcloth of these ideas that Dennett aims to meet 'the challenge of constructing a theory of mental events', using the data that scientific method permits. 'Such a theory will have to be constructed from the third-person point of view, since all science is constructed from that perspective' (CE 70). The method by which he proposes that this should be done he has dubbed 'the heterophenomenological method'. This consists in the 'experimenter' letting 'subjects' recount their thoughts, beliefs and feelings, their perceptions and memories, their likings and dislikings, their wishes, wants and intentions. The experimenter's raw data are mere noises emitted by organisms. But he must adopt the intentional stance, and 'treat the noise emitter as an agent, indeed a rational agent, who harbours beliefs and desires. . . the uttered noises are to be interpreted as things the subjects wanted to say, of propositions they meant to assert, for instance, for various reasons' (CE 76). The subject's self-ascriptions of experience are held to constitute 'a text': a fiction on a par with a novel. This alleged fiction is referred to as the subject's 'heterophenomenological world'. iy It is held to be a 'stable, intersubjectively confirmable, theoretical posit' (CE 81), having the same status as fictional objects. In the subject's heterophenomenological world, Dennett explains, there are intentional objects, to which various things happen. But, he claims, these objects, like Mr Pickwick, are made of nothing — they are fictional objects. But they are also said to be abstracta — not idle fictions, but 'hardworking theorists's fictions' (CE 95f.)- These fictional worlds are also claimed to be populated by 'all the images, events, sounds, smells, hunches, presentiments and feelings the subject sincerely believes to exist in his . . . stream of consciousness' (CE 98). The subject is not authoritative about what is happening 'in him', 'but only about what seems to be happening in [him]' (CE 96). Having extracted this 'heterophenomenology', Dennett explains that theorists can then turn to the question of what might explain the existence of this heterophenomenology in all its details. The heterophenomenology exists - just as uncontroversially as novels and other fictions exist. People undoubtedly do believe that they have mental images, pains, perceptual experiences, and all the rest, and these facts about what people believe, and report when they express their beliefs - are phenomena any scientific theory of the mind must account for. . . the question of whether items thus portrayed exist as real objects, events and states in the brain . . . is an empirical matter to

It is unclear why it is not deemed to be the subject's 'autophenomenological' world, and the experimenter's 'heterophenomenological' one.

428

Appendix

1: Daniel

Dennett

investigate. If suitable real candidates are uncovered, w e can identify them as the longsought referents of the subject's terms; if not, w e will have to explain w h y it seems to subjects that these items exist. (CE 98) 20 This p r o g r a m m e of research is, in our view, a non-starter, and the 'heterophenomenological method' incoherent. First, D e n n e t t wishes to construct 'a t h e o r y of mental events' by means of his h e t e r o p h e n o m e n o l o g i c a l m e t h o d . B u t it is unclear w h a t is supposed to c o u n t as a mental event. M a n y of the things h e m e n t i o n s are n o t events at all, and m u c h of w h a t he speculates a b o u t c a n n o t b e said to b e mental. Is seeing the tree i n the quad a mental event? Is hearing that the g o v e r n m e n t has fallen a mental event? Are loving Maisy a n d respecting Daisy mental events? K n o w i n g , believing, supposing, assuming, m e a n i n g s o m e t h i n g by w h a t o n e said, h o p i n g and fearing are clearly n o t m e n t a l events, since they are n o t eventsFeeling cheerful or depressed are m o o d s — that is, mental states — n o t events. Calculating in one's head a n d talking to oneself in one's imagination are mental activities or processes, n o t mental events. Is calculating o n paper or p r o v i n g a t h e o r e m aloud before an audience a mental event? It is doubtful w h e t h e r there is any c o h e r e n t subject matter here that could b e t h e subject of scientific investigation. 2 1 So it is equally unclear w h e t h e r there can be such a t h i n g as a. general theory o f ' m e n t a l events'. T o b e sure, there can b e n o such a thing as a general theory of n o n - m e n t a l events. Secondly, it is unclear w h o is supposed to u n d e r t a k e the challenge that D e n n e t t envisages. Is it m e a n t to b e a research p r o g r a m m e for cognitive neuroscience? O r for experimental psychology? O r for an imaginary 'science of consciousness'? T h e 'experim e n t e r ' w h o is supposed t o take tape recordings o r transcripts o f experimental subjects' ' h e t e r o p h e n o m e n o l o g y ' is supposed to investigate ' w h a t m i g h t explain the existence of this h e t e r o p h e n o m e n o l o g y in all its details'. So, w e m a y suppose that subject A declares that he is tired, that h e has a t o o t h a c h e , that h e is pleased that h e w o n his b e t w i t h Jack, that he believes that the R u r i t a n i a n Party will w i n the election, and that h e is afraid the e c o n o m y is going into recession. B u t one does n o t need a scientist to explain the existence of this ' h e t e r o p h e n o m e n o l o g y ' . A is tired, because h e has w o r k e d hard ail day long; h e has a t o o t h a c h e because there is a cavity in his t o o t h ; h e is pleased to have w o n his bet, because h e w a n t e d M y Love to w i n the 3.30; h e believes the R P will w i n the election because h e has seen the latest o p i n i o n polls; and h e is afraid that the e c o n o m y is g o i n g into recession because h e just read the financial n e w s . T h e s e are explanations of the existence of A's

20

Elsewhere Dennett elaborates: 'if we were to find real goings-on in people's brains that had enough of the "defining" properties of the items that populate their phenomenological worlds, we could reasonably propose that we had discovered what they were really talking about - even if they initially resisted the identifications. And if we discovered that the real goings-on bore only minor resemblance to the heterophenomenological items, we could reasonably declare that people were just mistaken in the beliefs they expressed, in spite of their sincerity' (CE 85). 21

Although we deny that the concept of the mental is obsolete (see below, p. 450), we agree with Searle to the extent that we think that it is quite useless for the kinds of purposes for which Dennett invokes it.

Appendix 1: Daniel Dennett

429

'heterophenomenology', and they need no science. What might call for science, more specifically for psychiatry, is why, for example, someone persists in believing something (e.g. that he has a terrible disease, or that everyone is 'out to get him') when all the evidence shows the groundlessness of his belief. Such abnormalities call for a psychiatric explanation (an explanation of why things are awry). But, of course, it does not follow that we need a comparable psychiatric (or any other 'scientific') explanation of why we believe well-founded beliefs. Thirdly, the supposition that the raw data for a future science of consciousness are noises that are to be interpreted as speech is misconceived. Human utterances are no more mere noises which stand in need of an 'interpretation' than the sights we see are mere patches of colour and shapes that need to be interpreted as the multicoloured objects that surround us. It is mistaken to suppose that a respectably scientific investigation into human psychology must take its raw data to be the bare noises of human speech (this would be to confuse phonetics with psychology). For (a) the raw data which we are given consist of significant human speech, not of bare sounds (phonetics is an abstraction from what is given.); and (b) there is no such thing as interpreting bare noises, any more than there is such a thing as interpreting a coded message (one can only interpret the message after it has been deciphered). Fourthly, the heterophenomenological method treats human avowals and reports of knowledge, belief, memory, perception, hopes and fears, pleasures and pains as fictions (like sentences in a novel). But this involves important inconsistencies. (i) To treat a text as fiction is, roughly speaking, to treat it as if it were preceded by a 'Once upon a time'. If the 'theorist' reads a transcript and treats it as fiction, he cannot also treat it as a theoretical posit. 'Once upon a time' does not introduce theoretical posits it precludes them. (ii) The alleged heterophenomenological world is held to consist of intentional objects, to which various things happen. But that makes no sense, since intentional objects such as that Maisy will come tonight, which is what A expects, or victory, which is what he hopes for, or treachery, which is what he fears, are not objects to which anything could possibly happen. (iii) Intentional objects such as that Maisy will come tonight are not fictions. Mr Pickwick, of course, is a fiction - a fictional character. It is not true that Mr Pickwick was made of nothing. Hamlet's father's ghost (also a fictional character) was made of nothing, but Mr Pickwick was made of flesh and blood. That Maisy will come tonight, unlike Mr Pickwick, is indeed not made of anything — but that does not make an intentional object much like a fictional ghost. (iv) The subject, according to Dennett, is not authoritative about the denizens of his heterophenomenological world ('what is happening in him'), but only about what seems (to him) to be happening within him. But that makes scant sense. First, if the subject avers that he believes, hopes, fears, expects this, that or the other, he is not reporting on anything that happens in him. Beliefs, hopes, fears and expectations don't 'happen in' people. But he does enjoy a prima facie authority on what he believes,

430

Appendix í: Daniel Dennett

hopes, fears and expects. His sincere avowal is a (defeasible) criterion for his so believing hoping, fearing or expecting. In the absence of defeating evidence, his word goes. Secondly, with respect to an array of mental phenomena, there is no such thing as 'seeming', and for a subject to enjoy 'authority' with respect to such items has nothing to do with anything that merely seems to him to be the case. So, for example, there is no such thing as a person seeming to himself to be in pain (a fortiori no such thing as its seeming to him that he is in pain and his not being in pain); and what goes for pain goes for other sensations too (e.g. tickles, tingles). Similarly, it cannot seem to one that one has a mental image before one's mind, even though one has none. (v) Dennett holds that there is a question as to whether the (intentional) denizens of people's heterophenomenological world exist as real objects, events and states in the brain. If not, then they exist only as fictions, and, he claims, we will have to explain why it seems to their subjects that they exist. This is confused. First, it surely does not seem to anyone that such intentional objects 'in the subject's heterophenomenological world' as that Maisy will come to dinner exist as 'real objects, events or states in the brain'. Indeed, it is unclear what, if anything, could be meant by asking whether an intentional object exists in the brain or as an object, event or state of some kind in the brain. The very notion of an intentional object was introduced by Brentano as something which need not exist at all in order to be the object of an intentional attribute such as belief or fear that something or other is thus and so. Secondly, it does not seem to people that their pains, beliefs, hopes and fears really exist. People really do have pains;22 they really believe, hope and fear things. There is no seeming about it — no matter whether these items can, or cannot, be identified with neural states or events. If they cannot be so identified, it does not follow that the pains are mere sham (fictions), so that we can close down all our hospitals; or that people do not really believe, fear and hope the various things they say they believe, fear and hope. Thirdly, Dennett confuses the intentional object of belief, hope or fear with the believing, hoping or fearing that is or may be intentional {and has an intentional object if it is). He claims that it is an empirical question whether someone's believing, hoping ox fearing something or other 'exists' as an object, event or state in the brain. (Of course, these are not intentional objects at all, but intentional attributes.) H e holds that if we find neural states or events in the brain that bear enough similarity to believing, hoping, etc., then the scientist could declare that these are neural states or events. If not, then people are just mistaken in thinking that they have any beliefs, hopes, fears, etc. But this is absurd. The properties that one's believing can have are, for example, being passionate (if one believes passionately that such-and-such) or tentative (if one believes tentatively) or justifiable {if one believes justifiably, i.e. is justified in believing what one believes). One's believings are essentially individuated by their object - that is, by what it is that is believed to be the

22

It is striking that Dennett apparently conceives of pain not as localized sensations of varying degrees of intensity, but a matter of 'having one's life hopes, life plans, life projects blighted by circumstances imposed upon one's desires, thwarting one's intentions' (CE 448). "Would that analgesics were so readily dispensable.

Appendix i: Daniel Dennett

431

case. And one may believe correctly or incorrectly, truly or falsely, depending upon whether what one believes is so or not. But it is not an open (or a closed) empirical question whether neural states or events are passionate, tentative or justifiable, or whether they are essentially individuated by their intentional object (since they have none); or whether they are correct or incorrect, true or false (since there is no such thing as a correct or incorrect, true or false neural state.) Finally, it is wholly obscure what might be meant by the idea that neural states could 'bear enough similarity' to believing or hoping to warrant declaring believing and hoping to be brain states. How could a brain state bear any significant similarity to believing that the Batde of Hastings was fought in 1066 or to hoping for salvation? It is important to realize that there is more in heaven and earth than is dreamt of in Dennett's philosophy. In particular, he offers us too few alternatives. The rejection of a Cartesian mind substance does not leave one with the dilemma that believing is either a state of the brain or a fiction. It is human beings who believe things, not their brains; and it is not a fiction, but a truth, that human beings do actually believe myriad things. So, if believing is neither a brain state nor a fiction, what is it? That is a bad question, for believing is not an anything; that is, there is no superior genus under which believing may be iiluminatingly or fruitfully subsumed. (Why should there be? Our general categorial terms in psychology were not devised by a divine Linnaeus for the ciassificatory purposes of cognitive scientists.) But there is nothing mysterious about that. The concept of belief can be perfecdy well elucidated by connective analysis, by displaying the logico-grammatical similarities and differences between believing and opining, thinking, gathering, supposing, assuming, accepting, knowing, being sure, being doubtful and so on.

4 Consciousness Dennett's aim in his work on consciousness is to show how 'a genuinely explanatory theory of consciousness' can be constructed (CE 256). He rejects the philosophical conception of consciousness that he refers to as 'the Cartesian Theater' (an amalgam of Descartes and the British empiricists). In its place, he suggests an account that he calls 'the Multiple Drafts' model. According to this, 'all varieties of thought or mental activity are accomplished by the brain by parallel, multitrack processes of interpretation and elaboration of sensory inputs' (CE 111). Information entering the nervous system is under continuous 'editorial revision', and there are multiple channels in which special circuits try, in parallel pandemonium, to do various things, creating Multiple Drafts as they go (CE 252). We should, he claims, 'think of the brain as a computer of sorts', as an information-processing system (CE 433). And we should think of ourselves as the program that runs on our brain's computer (CE 430). The central 'hypothesis' that Dennett defends is that 'Human consciousness is itself a huge complex of memes (or, more exactly, meme-effects in brains) that can best be understood as the operation of a "von Neumannesque" virtual machine implemented in the parallel architecture of a brain' (CE 210). (We shall explain what he means by 'memes' below.) Anything that has such a virtual machine as its control system 'is conscious in the

432

Appendix Í: Daniel Dennett

fullest sense of the word, and it is conscious because it has such a virtual machine 1 (CB 281). According to Dennett, 'the concepts of computer science provide the crutches of ;'-!-*^ imagination we need if we are to stumble across the terra incognita between our phenom- :? enology as we know it by "introspection" and our brains as science reveals them to us fc By thinking of our brains as information-processing systems, we can dispel the fog discovering how it might be that our brains produce all the phenomena' (CE 433). f j e has, he claims, explained the phenomena of human consciousness 'in terms of the operation of a "virtual machine", a sort of evolved (and evolving) computer program that - r e shapes the activities of the brain' (CE 431). : Having announced that his aim was to show how to construct a 'theory of consciousness', and having claimed that he has actually explained the phenomena of consciousness, on the very last page of his book Dennett observes that all that he has really done 'is to : replace one family of metaphors and images with another, trading in [Cartesian and v. empiricist metaphors] for Software, Virtual Machines, Multiple Drafts, a Pandemonium of Homunculi' (CE 455). We think that this is a perfecdy accurate summary of his achievement. Moreover, as we shall argue, the metaphors are poor ones, and serve no useful V; \ purpose in either neuroscientific discovery and theory or in philosophical clarification of the concept of consciousness. The Multiple Drafts model amounts to little more than a series of metaphors concerning largely unknown neural processes. Neither the brain nor its parts engage in any 'editorial processes'; there are no texts to edit in the brain, and the various concurrent neural processes occurring in the brain at any given moment are not in the least like a series of drafts of a text. The various parts of the cortex that are involved in, for example, perception are not akin to more or less stupid homunculi, and there is nothing intentional about these neural processes. For parts of the cortex are not subjects of intentional attributes. The suggestion that we should think of ourselves as computer programs is not coherent. Human beings are animals of a certain kind. They weigh so-and-so many kilograms, are of such-and-such a height, are either male or female; they are born, grow, fall in love, get married and have children, and so forth — none of which can intelligibly be said of computer programs. Perhaps Dennett means that our selves are computer programs — but that is no less absurd, for reasons we have explained (§12.4). The Cartesian self, understood as an immaterial substance, was a misconception; but that misconception is not rectified, but only exacerbated, by equating the 'self with a computer program. The suggestion that human consciousness is a complex of meme-effects in brains that constitutes the operation of a von Neumarmesque virtual machine implemented in the parallel architecture of the brain is simply unintelligible. We can make little sense of the thought that there is a von Neumannesque virtual machine implemented in the parallel architecture of the brain. That many neural processes run parallel with each other is no warrant for assuming that the brain is a parallel processing computer, let alone that it is implementing a program run by a serial, von Neumann computer. We can find no argument in Dennett's writings to render intelligible, let alone to provide evidence for, such a bizarre suggestion. We shall focus instead upon the incoherence of Dennett's notion of a meme and its role in his account of consciousness.

Appendix 1: Danie! Dennett

433

The term meme is derived from the writings of Richard Dawkins. 23 Memes are held to be complex ideas, which are readily memorable, and which 'replicate themselves' (just like genes) 'with reliability and fecundity' (CE 201). Examples of memes are such ideas as 'arch, wheel, wearing clothes, vendetta, right triangle, alphabet, calendar, the Odyssey, calculus, chess, perspective drawing, evolution by natural selection, impressionism, "Greensleeves", deconstructionísm'; 24 but also faith, tolerance, free speech, the conspiracy theory (DDI 349), and cooperation, music, writing, education, environmental awareness, arms reduction, The Marriage of Figaro, returnable bottles, and the SALT agreements, antiSemitism and spray-can graffiti (DDI 363). Once our brains 'have built the entrance and exit pathways for the vehicles of language', Dennett explains, 'they swifdy become parasitized (and I mean that literally, as we shall see) by entities that have evolved to thrive in just such niches: memes' (CE 200). Elsewhere, Dennett contends that the mind is a 'meme nest' (DDI 349) in which memes find a temporary home (DDI 355). Memes compete to enter human minds, where they take up residence (CE 203) and come into contact with each other (DDI 355). When they come into contact with each other, they adjust to each other, changing their phenotypical effects to fvt the new circumstances, and it is the recipe for the new phenotype that then gets replicated when the mind broadcasts or publishes the results of this mixing (DDI 355). Once our brains have been parasitized by memes, the memes transform the operating system or computational architecture of the brain (DDI 343), turning it into a mind (CE 252). Dennett compares the invisibility of genes with the invisibility of memes (ideas), the 'vehicles' of genes (organisms) with what he takes to be the 'vehicles' of memes (pictures, books, utterances (DDI 347)), the reproduction of genes with the alleged reproduction of memes, the phenotypic expression of genes with the alleged phenotypic expression of memes, the causal effects of genes with the alleged causal effects of memes. The notion of a meme is, as we shall show, incoherent. (i) The term 'idea' is confusingly multivalent. An 'idea' may be a concept, a thought, a proposition, a notion or a conception. It is, therefore, hardly surprising to find Dennett becoming entangled in his own terminology. For, while memes were introduced as 'complex ideas' which people communicate to each other, Dennett claims that such 'memes' as cooperation, music, education, The Marriage of Figaro and the Odyssey, not to mention the SALT agreements, 'are, all things considered, good from our perspective'

23

According to Dawkins, 'Examples of memes are tunes, ideas, catch phrases, clothes fashions, ways of making pots or of building arches. Just as genes propagate themselves in the gene pool by leaping from body to body via sperm or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain via a process which, in a broad sense, can be called imitation. If a scientist hears or reads about a good idea, he passes it on to his colleagues and students. He mentions it in his articles and his lectures. If the idea catches on, it can be said to propagate itself, spreading from brain to brain' (R. Dawkins, The Selfish Gene (Oxford University Press, Oxford, 1976), p. 206). 24 D. Dennett, Darwin's Dangerous Idea (Penguin, Harmondsworth, 1996), p. 344. Subsequent references in the text to this book will be flagged 'DDF.

434

Appendix 1: Daniel Dennett

(DDI 363) - and it is evident that he does not mean that the idea of The Marriage of Figaro or of the Odyssey is 'good from our perspective'. He observes that other memes are controversial, but should be tolerated - for example, television advertising. Should we tolerate the idea or the advertising? Other memes, he claims, are pernicious, but extremelv hard to eradicate - for example, spray-can graffiti, but is it the graffiti or the idea of such graffiti that is difficult to eradicate? The invitation to confusion evidently proved irresistible (ii) Genes are invisible in so far as they are molecules too small to be seen. Ideas concepts, propositions and conceptions are not too small to be seen. (iii) if organisms (as opposed to molecules) are vehicles of genes, then utterances pictures, books and artefacts are not vehicles of ideas - but human beings are. Utterances pictures, etc. are the means whereby we communicate ideas one to the other — hence they are analogous not to organisms, but to sperm and ova. But the analogy is poor. (iv) The distinction between genotype and phenotype has no intelligible application to ideas, concepts, propositions, etc. There is no analogue of statistical correlations between variations in genotype and variations in phenotype, or of the genuine explanatory force which these correlations have. (v) Ideas, concepts, propositions and conceptions, unlike genes, have no causal, but only logical, powers. But, according to Dennett, memes manipulate us (CB 203); they take up residence in our brains, shape its tendencies (CE 252), and transform its computational architecture (DDI 343). One may concede that acquiring ideas fin one or other of the numerous senses of this hopelessly ambivalent term) will affect various synaptic connections in the brain — but the idea acquired cannot. Yet it was supposed to be the idea that is the meme. (vi) To communicate an idea, a concept, a proposition or a theory to another human being is not to replicate anything. Einstein had a very good idea: he showed that E = mc2, and he communicated this to other physicists. This did not multiply the number of ideas in circulation, only the number of people acquainted with the very same idea. Communicating an idea to other people, fortunately, is not at all like impregnating them. There are thousands of people who know Einstein's equation, but there are not thousands of Einstein's equations - only one. And the proposition that E = mc 2 undergoes no propagation by being communicated to thousands of people - they all understand the very same proposition. So, it is altogether unclear what replication of memes is supposed to be. If the Odyssey is a meme, is its replication the copies of it that are produced by printers? But they do not exist in people's minds, or indeed in their brains; they exist in libraries and bookshops. And copies of books do not replicate themselves — if they did, we should have no need for printers. Is it the idea of the Odyssey that is supposed to be a meme? If so, what exactly is this idea? Is it simply knowing what the Odyssey is? But surely, knowing what something is, is not a meme! If a thousand people know what the Odyssey is, no idea has replicated itself. And if a thousand people all read the Odyssey (no matter whether it is the same copy or a thousand different copies), neither the Odyssey nor an idea of the Odyssey has replicated itself. We conclude that the alleged analogy between complex, memorable ideas and genes is a very poor one. It certainly is not strong enough to support an evolutionary theory of the 'survival of the fittest' ideas, parasitizing the human brain.

Appendix 1: Daniel Dennett

435

The account that Dennett offers is equivocal through and through. Memes, he claims, occupy minds, which are 'nests' or 'temporary homes' for them. Presumably this means that people acquire ideas, have ideas, believe such-and-such ideas. But belief and other intentional attributes that have ideas as their intentional objects are, according to Dennett, merely instrumentalist predictive devices on a par with centres of gravity or parallelograms of forces. It is far from clear how ideas can find a home in anything that has the ontological status of a centre of gravity. O n the other hand, memes 'take up residence in an individual brain, shaping its tendencies and thereby turning it into a mind' (CE 252). But, presumably noting the incoherence of supposing that there are any ideas in the brain, Dennett cautions that 'human consciousness is itself a huge complex of memes (or, more exactly, meme-effects in brains)' (CE 211, our italics). So it is not, after all, memes that take up residence in the brain, but only their effects — or rather, the effects of having an idea communicated to one. But we were told that it is memes that literally parasitize the brain, thereby turning it into a mind! We can see no hope of extracting any coherent account from this confusion. And the confusion fatally affects what purports to be an explanation of consciousness. The suggestion that consciousness is a complex of meme-effects in the brain is incoherent. Memes, we have been told, are complex ideas. Meme-effects in the brain are presumably meant to be the neural effects of acquiring certain complex ideas. Being awake, as opposed to being unconscious, and being conscious of something or other (e.g. of an object in one's perceptual field) are certainly not neural effects of the acquisition of ideas such as the idea of the wheel, of wearing clothes, of the Odyssey or of'Greensleeves'. But states of consciousness - that is, mental states that one may be in while conscious, such as being cheerful or sad, being in pain, or enjoying oneself— are not complexes of neural effects of ideas either. The claim that human consciousness is a complex of meme-effects in brains that can best be understood as the operation of a von Neumannesque virtual machine implemented in the parallel architecture of the brain is, we suggest, quite literally meaningless. We have argued that Dennett's account of intentionality rests on an inadequate conception of the logical character of the intentional phenomena to which Brentano drew attention, that it involves the misguided attribution of intentional attributes to kinds of things that could not logically possess any intentional attributes or behave as if they did. We have given reasons for thinking that Dennett's conception of an 'intentional stance' is confused, and that his claim that our normal employment of our psychological vocabulary is an instrumentalist predictive device on a par with such abstracta or theoretical posits as centres of gravity is incoherent. We have tried to demonstrate that the idea that a new science of consciousness, dependent upon the heterophenomenological method, is about to emerge is chimerical. For, we argued, the 'heterophenomenological method' is worthless. Finally, we have challenged Dennett's 'Multiple Drafts model' and his claim that consciousness is a huge complex of meme-effects in brains that can be best understood as the operations of a 'von Neumannesque' virtual machine implemented in the parallel architecture of the brain. If our arguments hold, then Dennett's theories of intentionality and of consciousness make no contribution to the philosophical clarification of intentionality or of consciousness. Nor do they provide guidelines for neuroscientific research or neuroscientific understanding.

Appendix 2

John Searle

John Searle has written extensively about philosophical problems that impinge upon cognitive neuroscience. In his 1984 Reith Lectures Minds, Brains and Science he addressed the lay public in a series of six lectures on topics that are relevant to our discussions in this book. These lectures were delivered with his customary panache and clarity. In The Rediscovery of the Mind (1992), writing primarily (but not exclusively) for professional philosophers, he confronted the large theme of consciousness. In that book he advocated his conception of biological naturalism, and criticized various misconceived doctrines that have dominated philosophy of mind, psychology and cognitive science in the twentieth century, such as behaviourism, identity theory, 'black box' functionalism, strong AI (i.e. 'Turing machine' functionalism) and eliminative materialism. Between 1995 and 1997 he contributed a series of critical reviews to the New York Review of Books of various books on consciousness, including works by Crick, Edelman, Penrose, Dennett, Chalmers and Rosenfeld. These reviews were published in book form in 1997 in The Mystery of Consciousness. In the 'Millennium' edition of The Philosophical Transactions of the Royal Society (vol. 354, 1999), he addressed scientists in general, and neuroscientists in particular, on the subject of 'The future of philosophy'; delineated the relationship between philosophy and science as he sees it; and, among other things, outlined his conception of the traditional mind—body problem and of the relations between philosophy of mind and cognitive science. Searle's work is widely read by cognitive neuroscientists. It is admired for its lucidity and forceful argument. It addresses a multitude of issues that are of concern to them, and displays familiarity with current advances in cognitive neuroscience. Since we have disagreed with Searle on a variety of points, it is important to make explicit the rationale for differences between the conception we have advocated in this book and the views that Searle has defended in his writings. We shall focus upon methodological issues concerning the nature of philosophical investigation, the methods of philosophy, the boundaries between philosophy and science, and upon substantive issues pertaining to the philosophical investigation of consciousness.

1 Philosophy and Science According to Searle, there is no sharp dividing line between science and philosophy. In this respect, Searle works within the received American tradition that has been influenced so much by W. V. O. Quine.

Appendix 2: John Searle

437

Both science and philosophy, according to Searle, are (a) universal in their subject matter and (b) aim at truth. Nevertheless, he asserts, there are important differences. Philosophical problems tend to have three related features that scientific problems lack. First, philosophy is in large part concerned with questions for the answering of which there is as yet no satisfactory and systematic method. Secondly, philosophical questions tend to be what Searle calls 'framework questions' - that is, questions that deal with large frameworks of phenomena, rather than widi specific individual questions. Thirdly, philosophical questions are typically about conceptual issues.1 The first and second points are meant to explain why philosophy stands in a rather peculiar relation to science. For, Searle claims, as soon as we can revise and formulate a philosophical question to the point where we can find a systematic way of answering it, it ceases to be philosophical and becomes scientific (so, for example, the debate between vitahsts and mechanists was rendered obsolete once we came to understand the molecular basis of living organisms). And he envisages something similar happening in the current debate concerning the nature of consciousness and its relation to brain processes. This conception of the relationship between philosophy as the clarifier of confused questions and of science as the answerer of clear questions explains, according to Searle, why science appears to have such a good record of success and philosophy seems to be an irremediable failure. For, as soon as we find a systematic way to answer a question, and get an agreed answer from competent investigators, we cease calling the problem 'philosophical' and start calling it 'scientific'. So, philosophy cannot answer, solve or resolve questions, but only contribute to their clarification. Once it has done so, the question becomes a scientific one and is answered by science. It should be clear that the conception of philosophy defended by Searle is very different from the one advocated and implemented in this book. We hold that there is a radical dividing line between philosophy thus practised and the natural sciences. We agree with Searle that philosophical questions are typically about conceptual issues, and we hold that philosophical problems and puzzles are typically the result of conceptual entanglement or of one form or another of conceptual unclarity. These problems and puzzles cannot be solved, resolved oc dissolved by scientific theory or by scientific experiment. Whatever light we have succeeded in shedding upon the conceptual problems and puzzles that patently concern cognitive neuroscientists has been the result of a priori connective analysis, not of scientific experiment or empirical theory. The natural sciences aim to explain contingent phenomena of nature. They construct theories in terms of which the phenomena can be explained, typically hypotheticodeductively. The theories and hypotheses are confirmed or disconfirmed by observation and experiment. The predictions the validation of which confirm a scientific theory may be only approximations to the truth, the degree of acceptable approximation varying with subject matter and with the standards of accuracy of the science of the time. The natural sciences are typically hierarchical and progressive, the positive results achieved

1

John R. Searle, 'The future of philosophy', Philosophical Transactions of the Royal Society, B 354 (1999), p. 2069. 2 Ibid., p. 2070.

438

Appendix

2: John

Searle

c o m m o n l y b e i n g the basis for t h e construction o f y e t m o r e c o m p l e x , m o r e encompassing theories. B y contrast, the connective-analytic enterprise of philosophy d o e s n o t aim to explain natural p h e n o m e n a at ail. It is n o t a form of armchair science, that seeks, w i t h o u t the l a b o u r o f e x p e r i m e n t a n d observation, t o find answers to t h e questions that are t h e p r o p e r p r o v i n c e o f t h e sciences. I n the sense in w h i c h t h e natural sciences explain p h e n o m ena, analytic philosophy explains n o t h i n g . R a t h e r , it describes conceptual structures a n d connections, disentangles conceptual confusions, a n d explains, w i t h o u t a n y theory o r hypothetico-deductive explanation, w h y w e b e c o m e entangled in t h e w e b o f concepts that w e d e p l o y in o u r t h i n k i n g a n d theorizing a b o u t the w o r l d , a n d so fall i n t o certain kinds o f confusion. I n t h e sense i n w h i c h there are theories in t h e natural sciences, there are n o licit theories i n philosophy (any m o r e than t h e r e are, in this sense, theories in mathematics). T h e r e are n o hypotheses i n philosophy, n o r is there anything hypothetical a b o u t t h e answers w h i c h analytic philosophy offers to its problems: there is n o 'if suchand-such, t h e n things will t u r n o u t t h u s - a n d - s o - as will b e observed in an e x p e r i m e n t ' . A n d of course, there are n o experiments, n o experimental confirmation or disconfirmation of theories, and n o predictions the truth or falsity o f w h i c h m i g h t validate or falsify a theory. T h e procedures of philosophy thus understood are w h o l l y a priori, and its p r o p o s e d solutions or resolutions o f its problems are validated or refuted b y a priori a r g u m e n t , as are t h e p r o b l e m s o f mathematics. 3 B u t w h i l e mathematics is a m a t t e r o f concept formation b y constructing proofs, philosophy is a m a t t e r o f c o n c e p t elucidation. M a t h e m a t i c s is a synthetic, constructive activity, creating n e w conceptual c o n n e c t i o n s , whereas philosophy is an analytic, descriptive activity, clarifying a n d elucidating existing conceptual connections. T h e p r o p e r task o f the natural sciences is to d e t e r m i n e h o w things are in t h e w o r l d a n d w h y they are thus or otherwise. It is therefore c o n c e r n e d w i t h w h a t is true and w h a t is

3

It is widely believed by American philosophers that W. V. O, Quine, in his 'Two dogmas of empiricism' (Philosophical Review, 60 (1951) pp. 20-43), showed that the customary distinction between analytic and synthetic propositions is not viable, and hence too that the different distinction between a priori and empirical propositions is equally untenable. Consequently, he is thought to have shown that one cannot demarcate philosophy from science by reference to philosophy's concern with a priori questions in contradistinction to science's concern with empirical fact. But this is mistaken. At most Quine showed that the conception of analyticity propounded by his teacher Rudolf Carnap is untenable (although Carnap denied that he had succeeded in so doing). But even if Quine was right and Carnap wrong, the Camapian conception of analyticity is not the only one; moreover, Quine never responded adequately to the powerful reply to laim by P. F. Strawson and H. P. Grice which did articulate a defensible conception of analyticity in the face of Quine's criticism ('In defence of a dogma', Philosophical Review, 65 (1956), pp. 141-58). Furthermore, one may reject the notion of analyticity altogether, as "Wittgenstein did, and still cleave to the distinction between what is a priori and what is empirical (see P. M . S. Hacker, Wittgenstein's Place in TwentiethCentury-Philosophy Analytic ch. 7, and H.-J. Glock, 'Wittgenstein vs. Quine on logical necessity', (Blackwell, Oxford, 1996), in S. Teghrarian (ed.), Wittgenstein and Contemporary Philosophy (Thoemmes Press, Bristol, 1994), pp. 185-222).

Appendix 2: John Searle

439

false, and its successes contribute to our knowledge about the world. Philosophy, by contrast, is not concerned with empirical truth and falsehood, but rather with questions of sense. It is concerned with the determination of the bounds of sense, and with the explanation (by description) of the ways in which they are transgressed, both in scientific conjecture (including parts of cognitive neuroscience that have been our concern) and in philosophical reflection itself But there can be no approximations to sense (whereas there can be approximations to truth in scientific theory and prediction), for any deviation from sense is one form or another of nonsense. And there can be no theories that determine, with such-and-such margin of error, what does and what does not make sense. For there is no margin of error in the determination of sense. Analytic philosophy, in the domain of the philosophy of logic and language, metaphysics, epistemology and philosophy of psychology, 4 above all clarifies concepts and conceptual networks. It is concerned not with the description and explanation of empirical facts, but with the elucidation of the forms in which we describe empirical facts - that is, with the description of our conceptual scheme. It does not add to our knowledge of the world, but contributes to our understanding, in the face of very special kinds of confusions — namely, conceptual confusions - of the knowledge we already have. For its results are not, and cannot be, startling new facts and theories, but only the clarification of the forms of thought we employ, yet find exceedingly difficult to bring into focus - especially in domains in which there is, for various reasons, a perennial danger of conceptual entanglement. According to Searle, both science and philosophy are universal in their subject matter. 5 This, in our view, is mistaken. First, natural science is not universal in its subject matter: its domain consists of the phenomena of nature. There is no natura! science (or any other science) of literature or of art, although both are meet subjects for serious study. Nor is there a natural science of morality, politics or law, although these subjects are legitimate spheres for human investigation and for the achievement of knowledge. There is no science of history, although the study of history is pivotal to the humanities and to our understanding of the development of science. There are, of course, social sciences, but whether the methods and forms of explanation of the social sciences are akin to those of the natural sciences, whether we can hope to discover laws of human society on the model of laws of nature which the natural sciences aim to disclose, is a highly contentious issue. But even methodological monists who think that the social sciences may aspire to discover social laws should hesitate to conceive of the study of history as a domain of science. Secondly, it is less than obvious what might be meant by saying that philosophy is universal in subject matter, it is surely false that philosophy and empirical science compete

4

Our remarks henceforth will be restricted to these branches of philosophy. Our claims would require qualification if ethics and lega! and political philosophy were under consideration. For these are concerned not only with describing and clarifying our conceptual framework but also with reflecting on how we ought to live. 5 Searle, 'Future of philosophy', p. 2069.

440

Appendix 2: John Searle

over the very same (universal) subject matter. The natural sciences are restricted to the domain of nature, but philosophy, even philosophy of science, is surely not concerned with making empirical discoveries about nature, in competition with science. What is true is that there are domains that are very much of concern to philosophy which are of none to science. While there is no such thing as a science of history, there is philosophy of history; while there is no such thing as a science of art, there is such a thing as philosophy of art. But the question is: what is the nature of the concern that philosophy has within its universal domain? For philosophy of history is no more in competition with history than philosophy of physics is in competition with physics. The sense in which philosophy, unlike science, is universal is that its concern is characteristically with conceptual structures and relations within any domain of human thought and experience in which conceptual questions and confusions arise - no matter whether in the proper domain of the natural sciences or in the domain of the a priori sciences of mathematics, in the social sciences, history, the arts, morality, law, politics, etc. O n the other hand, it is misleading to suggest that philosophy is primarily concerned with 'framework' questions. It is true that among the questions of philosophy, there are many that are characterized by the highest degree of generality (e.g. what are the most fundamental objects of reference in our conceptual scheme? what are the conditions of the possibility of knowledge of objective particulars? what is the nature of necessity?). But it is unhelpful to suggest that philosophical questions can be characterized by their generality or by their concern with 'frameworks of phenomena'. There are endless very particular philosophical questions, with many of which we have been concerned in this book (e.g. how are knowledge how and knowledge that related? are mental images representations of what they are images of? is thinking an activity? is memory exclusively of the past? is a voluntary movement a movement caused by an act of will?). But one may happily concede to Searle that even the most particular philosophical question may be connected with, and ramify into, the broadest and most general of 'framework questions'. For we are moving within the web of words, examining the structure of our conceptual scheme, and - in a sense - everything is connected with everything else: each node within the web is linked with numerous adjacent nodes. So very specific questions lead immediately to more general ones. That is why local disagreements among philosophers rapidly ramify into global ones. Searle claims that both science and philosophy aim at the truth - and in a sense that is obviously correct. But it can be misleading. The natural sciences aim to discover truths about nature, to unify and explain them by means of ever wider and more powerful theories. The social sciences, such as economics and sociology, aim to discover truths about the functioning of human society or societies. Many of the truths that the sciences have discovered are, or were, exceedingly surprising and unexpected. This applies to the social (and psychological) sciences no less than to the natural sciences, irrespective of whether the logical forms of explanation in them are or are not the same. What are the truths that philosophy aims to disclose? They concern conceptual connections and relationships. These are normative — rule-governed connections within the web of words. So they are not so much truths about the world as truths about the modes in which we represent truths about the world to

Appendix

2: John Searle

441

ourselves. 6 M o r e o v e r , although philosophy m a y p o i n t out a c o n n e c t i o n (an implication, a presupposition, a compatibility or incompatibility) that had escaped notice in philosophical reflection or in scientific theorizing, there is an i m p o r t a n t sense in w h i c h this cannot b e a n e w discovery, b u t at most s o m e t h i n g w h i c h h a d n o t b e e n realized. It c a n n o t b e a g e n u i n e novelty, like a scientific discovery that yields n e w k n o w l e d g e . For if it were a novelty, t h e n it c o u l d n o t b e part of o u r existing conceptual scheme, b u t at most a modification of it that is being r e c o m m e n d e d . 7 If philosophy a i m e d at the discovery of n e w truths after the manner of the empirical sciences, it w o u l d b e puzzling that it has, in its long history, c o m e up w i t h so few results. Physics, chemistry and biology have a shorter history, b u t they can fill libraries w i t h accounts of t h e k n o w l e d g e they have achieved. It should b e puzzling, or alternatively very depressing, that philosophy has so little to s h o w b y w a y of comparable achievement. F o r although it can fill libraries, these libraries are n o t repositories of philosophical k n o w l e d g e . O n e explanation favoured b y some neuroscientists (see §14.4) is that the apparent poverty of the results of philosophy is to b e explained by reference to the fact that its practitioners (e.g. Descartes, L o c k e , H u m e , Kant, Russell, Wittgenstein) are i n c o m p e t e n t a n d their armchair m e t h o d s are inappropriate to their goal. H e n c e , as w e saw, s o m e neuroscientists t h i n k that philosophy should step aside and let neuroscience take o v e r the p r o b l e m s c o n c e r n i n g t h e m i n d , consciousness a n d self-consciousness w i t h w h i c h p h i l o sophers have b e e n struggling for the last three and a half centuries. B u t this explanation, w e argued, is predicated u p o n a fundamental misunderstanding of philosophical p r o b l e m s . Searle offers a quite different explanation. H e claims that philosophy is c o n c e r n e d w i t h 'questions that w e have n o t yet found a satisfactory and systematic w a y to answer'. I n his view, 'as soon as w e can revise a n d formulate a philosophical question to the p o i n t that w e can find a systematic w a y to answer it, it ceases to be philosophical and b e c o m e s scientific'. This is a v i e w that Russell advanced," and it was, in a m o r e qualified m a n n e r ,

b

Indeed, they are 'truths' in a Pickwickian sense, since they are, in effect, norms of representation, which we present to ourselves in the misleading form of descriptions of fact. They are not true norms of representation (there is no such thing); rather, it is true that they are our norms of representation, and are partly constitutive of their constituent concepts. 7 This is a complex and ramifying point. It turns on the claim that to possess a concept is to have mastered the use of a word (or phrase). T o have mastered the use of a word is to know how it is to be employed, hence to know what counts as correct and what as incorrect use. So it is to know (explicitly or implicitly) the rules for the use of the word, and to use the word in accordance with those rules. But there can be no hidden, unknown rules for the use of words, for rules are guides to conduct and standards of correctness for conduct, and there can be no hidden, unknown guides to (or norms of) conduct which inform the actions of participants in a practice. For, if they are hidden, then they cannot be used as standards of correctness by reference to which mistakes are identified and corrected, and in terms of which learners of the language are instructed in the correct use of terms. There can be hidden regularities in normative, rule-governed practices - i.e. regularities of which the participants are wholly unaware - but there can be no hidden rules. s In 1912, in his introductory book The Problems of Philosophy (Oxford University Press, Oxford, 1967), p. 90, Russell wrote, that 'as soon as definite knowledge concerning any subject becomes possible, this subject ceases to be called philosophy, and becomes a separate science'.

442

Appendix

2: John Searle

advocated by Searle's teacher, Austin. 9 It is a c o n v e n i e n t w a y of explaining w h y there is n o philosophical knowledge (comparable to the k n o w l e d g e achieved b y physics) a n d w h y there are n o established philosophical theories (comparable to theories in chemistry) w o r t h speaking of. B u t this c o n c e p t i o n can and should be challenged. It is perfectly true that t h r o u g h o u t its long history the t e r m 'philosophy' has e n c o m passed a m u l t i t u d e of different kinds of subject. It was, after all, only relatively recently, that physics ceased to b e k n o w n as natural philosophy'. In the seventeenth century, a great philosopher, Descartes, did n o t sharply distinguish his philosophical investigations from his scientific ones, v i e w i n g his Optics, for example, as a straightforward application of t h e m e t h o d s advocated in his Discourse on Method; a n d a great scientist, B o y l e , did n o t distinguish his empirical investigations i n t o the nature of matter from his metaphysical reflections o n the ontology of secondary qualities. Physics split off from philosophy in the late seventeenth century, a n d t h r o u g h o u t the e i g h t e e n t h a n d n i n e t e e n t h centuries there was a tendency to conceive o f p h i i o s o p h y as an investigation i n t o the nature of the h u m a n m i n d . B u t at the e n d of the n i n e t e e n t h a n d b e g i n n i n g of t h e t w e n t i e t h century, psychology split off from philosophy, and b e c a m e an i n d e p e n d e n t science. D o e s this s h o w that as soon as w e can revise and formulate a philosophical question to the point at w h i c h w e can find a systematic way to answer it, it ceases to be philosophical and b e c o m e s scientific? N o . It merely shows that philosophy's perennial craving for a special, first-order subject matter has repeatedly b e e n frustrated. T h a t in t u r n shows t w o things. First, that t h r o u g h o u t t h e history of w h a t is called 'philosophy', empirical questions w e r e c o m m o n l y i n t e r w o v e n w i t h non-empirical, conceptual ones. O n c e they w e r e sharply distinguished, the empirical questions m o v e d into the p r o v i n c e of science. T o that extent, Searle is correct. Secondly, it shows that the quest for a first-order subject m a t t e r is futile. W h a t remain w i t h i n the province o f p h i i o s o p h y , correctly conceived, are conceptual questions (and c o n ceptual confusions) that are n o t amenable to scientific, experimental treatment at all. It is n o t true that as soon as w e can find a systematic w a y of answering these questions, they are ipso facto transformed into scientific questions - h e r e Searle is quite mistaken. A l t h o u g h questions a b o u t t h e empirical n a t u r e o f material objects a n d m a t t e r b e l o n g to the province of science, w h a t the logical features of substance concepts (of sortal n o u n s (such as ' m a n ' , ' d o g ' , 'cabbage') and concrete mass n o u n s (such as 'water', 'steel', 'oxygen')) are is n o t a question for empirical science, but for conceptual analysis. W h e t h e r the

9

J. L. Austin, Tfs and cans', repr. in his Philosophical Papers (Clarendon Press, Oxford, 1961), p. 180. H e wrote: 'In the history of human inquiry, philosophy has the place of the initial central sun, seminal and tumultuous: from time to time it throws off some portion of itself to take station as a science, a planet, cool and well-regulated, progressing steadily towards a distant final state. This happened long ago at the birth of mathematics, and again at the birth of physics: only in the last century we have witnessed the same process once again . . . in the birth of mathematical logic, through the joint labours of philosophers and mathematicians. Is it not possible that the next century may see the birth . . . of a true and comprehensive science of language? Then we shall have rid ourselves of one more part ofphiiosophy (there will still be plenty left) in the only way we ever get rid ofphiiosophy, by kicking it upstairs.'

Appendix 2: John Searle

443

identity of substances is absolute or merely relative identity is not a question that is likely to be resolved by physics. Moreover, the autonomy of physics has added new questions to the philosophical agenda - questions in the philosophy of physics concerning, for example, the logical nature of explanation in the physical sciences, the ontology of theoretical entities in physics, the reducibility or non-reducibility of other sciences and scientific explanations to those of physics. Similarly, the autonomy of the psychological sciences has not bankrupted philosophy of psychology. It is not as if empirical psychology can discover by experiment whether belief is a mental state or a mental disposition or neither. Neuroscience is not going to resolve questions about the concepts of consciousness and self-consciousness, any more than questions concerning the neural basis of intransitive consciousness are going to be answered by the a priori methods of philosophy. Searle claims that 'as soon as we find a systematic way to answer a question, and get an answer that ail competent investigators in the field can agree is the correct answer, we stop calling it "philosophical" and start calling it "scientific'". 10 This is questionable, on two counts. First, it implies that there are no systematic ways of answering questions in philosophy. This is wrong. One may indeed claim that there is no single 'universal method', any more than there is in science. But there are methods, which we have exemplified again and again throughout this book. For example, the meticulous examination of the grammar of problematic expressions, of the rules for the use of the words in question; the investigation of the logical implications of the use of such an expression in certain sentences, its entailments, compatibilities and incompatibilities; the scrutiny of the presuppositions of the use of the problematic expression, of the behavioural contexts in which its use may be embedded from occasion to occasion; the examination of its semantic field - that is, its relations to other expressions in the same domain; the clarification of the ways in which we can confirm or verify the application of the expression; and so on and so forth. Secondly, Searle's claim implies that once philosophical clarity with respect to any given question is achieved, the question ceases to be philosophical, and its answer is 'scientific'. But this is mistaken. Competent philosophers by and large agree that the five proofs of the existence of God propounded by Aquinas are all invalid - but does this make the disproofs 'scientific? Hume argued, contrary to many seventeenth-century thinkers, that causal knowledge is to be obtained by empirical observation and not by a priori argument, and that causal generalizations are to be established by inductive methods. Most philosophers (and scientists) agree. Does this imply that Hume's claim is part of an experimental science, established by observation and experiment?

2 Searle's Philosophy of Mind We are in complete agreement with Searle in his repudiation of Cartesian dualism and of behaviourism, identity theory, eliminative materialism and functionalism in its various

Searle, 'Future of philosophy', p. 2070.

444

Appendix 2: John Searle

forms. Searle characterizes his conception of consciousness as 'biological naturalism', bv which he. means that consciousness is a natural biological phenomenon, which is to be studied by the biological sciences. With this we have no quarrel. Although we distinguish philosophical, conceptual questions about consciousness from empirical ones, it is our view that there are many aspects of intransitive and transitive consciousness that are investigable by neuroscience, which can investigate the neural concomitants of consciousness in its various forms and the neural aetiology of various conscious states. Of course, the scientific discoveries will not solve or resolve the conceptual questions, just as the answers to the conceptual questions will not provide answers to the empirical ones. We part company with Searle, however, when he claims that 'mental phenomena are caused by neurophysiological processes in the brain and are themselves features of the brain'. 11 Consciousness, in his view, 'is a feature or property of the brain in the sense, for example, that liquidity is a feature of water'.' 2 The claim that consciousness 'is caused by lower-level micropro cesses in the brain' is misleading, but fairly harmless. It is misleading inasmuch as it conflates a causal condition with a cause.13 It is of course true that, but for a multitude of only partially understood processes in the brain, one would not be conscious at all, and that, but for a variety of even less well-understood processes, one would not be conscious of the various things of which one is transitively conscious. Equally, but for a variety of lower-level microprocesses in the brain, one would not walk or run, get up or sit down. Nevertheless, one would surely not say that the cause of A's running or of B's sitting down is a process in their brain. If asked what made A wake up (and, in this sense, regain consciousness), one might reply, 'The knock on the door', or 'The ringing of the telephone', but surely not 'The microprocesses in his brain'. Similarly, if one were asked what made one conscious of someone's being behind the curtain or of one's ignorance, one might reply, 'It was the , sudden movement of the curtain' or 'It was the speaker's remarkable display of erudition', but, again, not 'Why, microprocess M in the brain, of course'. For the latter are standing causal conditions, which we normally distinguish from the operative cause. The claim that consciousness is a feature of the brain, however, is more grievous. For, as we argued at length in chapter 3, to ascribe consciousness and experiences to the brain is to ascribe to a part of an animal properties which can intelligibly be ascribed only to the animal as a whole; that is, it is to commit the mereological fallacy in neuroscience. Consciousness is not, and could not be, a feature of a brain.14 Brains are no more conscious than they can go for walks or climb trees, even though it is true that an animal cannot go for walks or climb trees unless its brain functions appropriately. It is animals, including human beings, that are conscious or unconscious, that lose and later regain 11

j . R. Searle, The Rediscovery of the Mind (MIT Press, Cambridge, MA, 1992), p. 1. Ibid., p. 105. Of course, sometimes microprocesses in the brain may indeed be the cause, and not merely the causal condition, of some phenomenon of consciousness, e.g. of depression, or of the 'shattered mirror effect' of migraine. 14 Unless all that is thereby meant is that an animal will be conscious only if appropriate brain processes are going on. 12

Appendix 2: John Searle

445

consciousness, and that may become conscious of this or that if their attention is caught and held by some feature of their environment. They would not do so, of course, but for certain processes occurring in their brains, but it does not (and could not) follow that their being conscious is a feature of their brains. Liquidity can indeed be said to be a feature or property of quantities of water at appropriate temperatures. There are criteria for whether water is in a liquid, solid or gaseous state. But there are no criteria for whether a brain is in a conscious or unconscious state, any more than there are, or could be, criteria for whether a molecule of water is in a liquid, solid or gaseous state. There are no criteria for whether the brain is conscious of this or that, only inductive correlations between brain states and the animal's being conscious, unconscious or transitively conscious of this or that feature of its environment, or in one or another state of consciousness. According to Searle, the brain is a machine, a biological machine, and it can think. Human brains, he contends, sometimes compute; they add two and two and get four, for example. 15 Here too we part company with him. Whether brains are machines is a debatable matter that turns on the notion of a machine, which need not be debated here and is not decisive for our disagreements with Searle. However, our brains neither think nor compute — we do (although, of course, we would not be able to do so but for a variety of brain (and other) processes and states that make thinking and computing possible). The brain no more adds two and two to make four than does an abacus; it cannot intelligibly be said to possess the concept of addition or of identity (or any other concept), or to have grasped the use of numerals (or of any other symbol).16 As we have argued, for a being to be able to think (beyond the rudimentary thinking of non-human animals), it must also be able to be thoughtless and inconsiderate, or thoughtful, introspective or reflective. It must be able to think before it speaks, to speak without thinking, or to think while it is speaking. It must be capable of idle speculation or of purposeful reflection. It must be able to reconsider what it previously pondered, and so on and so forth, through myriad nuances which we distinguish with respect to the cogitative activities of human beings, but which we do not predicate of their brains. That we do not do so is not a mark of correctable ignorance - of failure to recognize the facts for what they are — but of our assigning no sense to the form of words 'My brain is thinking things over', 'His brain is in a pensive mood to day', or 'Her brain is thoughtless'. As we observed in §4.1, Searle holds that sensations such as toothache or stomach-ache occur not in one's teeth and gums or in one's midriff, but in the brain. He writes: 'Common-sense tells us that our pains are located in physical space within our bodies, that for example a pain in the foot is literally inside the area of the foot. But we now know that is false. The brain forms a body image, and pains, like all bodily sensations, are parts of the body image. The pain-in-the-foot is literally in the physical space of the brain.' Pains 'in the physical space of the brain' — that is, in the head - are headaches, not pains

15 ,G

j . K. Searle, The Mystery of Consciousness (Granta Books, London, 1997), p. 13. By the same token, of course, it cannot be said to lack the relevant concepts or to have failed to

grasp the use of numerals. Only what can acquire concepts can be said to lack certain concepts. 17 Searle, Rediscovery of the Mind, p. 63.

446

Appendix 2: John Searle

in the foot, and, of course, they are not in the physical space of the head as pins may be in the head.18 For pains are not kinds of objects that may be inside or outside physical objects, or that may be inserted into objects and then taken out again, and they are neither smaller nor larger than the part of the body in which they are located. For one to have a backache is simply for one's back to ache; to have a pain in one's foot is for one's foot to hurt, not one's brain or head. The criterion for the location of a pain is where the sufferer naturally assuages, where he points when asked where it hurts, arid where he says he feels the pain - that is not 'common sense', but logical grammar (§4.1). It is not an opinion, but an aspect of the meaning of the phrase 'pain location'. According to Searle, the phenomenon of pain in phantom limbs makes it evident that we experience bodily sensations in the body image. Moreover, 'many of us have a version of the phantom limb in the form of sciatic pain'.19 But the phenomenon of pain in a phantom limb shows that we can have the kinaesthetic illusion of possessing a limb that we have in fact lost — that is, it feels just as if the limb were still there — and we can feel a pain where the limb is felt to be; but it does not show that the pain is in the brain or in something called 'a body image' allegedly in the brain. Furthermore, this phenomenon is not comparable to the phenomenon of sciatic pains that are felt in the foot, although their cause is in the spine. Here it really is one's foot that hurts, whereas in the case of pain in a phantom arm it is not actually one's arm that hurts, since one has no arm. According to Searle, consciousness has what he calls 'a first-person ontology', by which he means that (i) conscious states exist only when experienced by a person, and (ii) they exist only from the first-person point of view of that person. The mark of conscious states, in his view, is that for any such state there is something that it qualitatively feels like to be in it. In this sense, all conscious states are qualitative, subjective experiences, and hence are qualia.20 This is confused (§§10.3-10.3.5). It is correct that there are no experiences that are not someone's experiences, no pains that are not someone's pains. If that gives conscious states a 'first-person ontology', well and good. But then it also gives smiles and sneezes a 'first-person ontology', for, the smile of the Cheshire cat apart, there are no smiles that are not someone's smiles and no sneezes that are not someone's sneezes. To be sure, smiles and sneezes cannot be said to exist 'only from the first-person point of view' of the smiler or sneezer — indeed, it is not clear what that claim would mean. Smiles and sneezes don't exist (or fail to exist) from any 'point of view', although they can be seen or heard by anyone in the vicinity. But it is equally unclear why Searle claims that conscious states exist only from a first-person point of view. It seems that his idea is that since every experience is necessarily someone's experience, therefore each person stands in a special relationship to his own experience. 21 That 'relationship', it seems, is conceived to ,Si

As we argued, the grammar of the phrase 'in the head' varies (without ceasing co be a spatial, locative phrase), depending on whether the subject term is a material object, such as a pin, or a sensation, such as a pain. For the logical consequences of the resultant propositions, their implications, compatibilities and incompatibilities vary accordingly. 19 20 21

Searle, Mystery of Consciousness, p. 182. Ibid., pp. xiv and 9. Searle, Rediscovery of the Mind, p. 95.

Appendix 2: John Searle

447

be a form of inalienable ownership, which guarantees the existence of the experience as long as this relationship obtains: for the experience, it seems, ceases to exist as soon as it ceases to be owned. If that is what Searle thinks (and we are not sure it is), then it is mistaken. To have a headache is not to stand in a relationship to a headache — it is for one's head to ache. To love Maisy may be said to be a matter of standing in a certain emotional relation (viz. of loving) to Maisy, but to have the experience of loving her is not a matter of standing in a further relationship to the experience of loving. Experiences are not mental 'objects' (just like material objects, only immaterial), since they are not 'objects' of any kind. In particular, they are not relata that stand in a relation (of being had) to the subjects of the experiences. To have an experience is not to own or possess anything, and experiences are not kinds of private property. One might perhaps say that experiences are attributes of subjects of experience, but then one must add the proviso that one no more stands in a relationship to one's own attributes than coloured objects stand in relation to their colours. Different people may have the very same headache or, more generally, pain (and that may betoken the fact they are suffering from the same disease). One cannot say, 'But they must be different, since yours is yours and mine is mine', for the subject of pain (i.e. being had by such-and-such a person) is not an identifying characteristic of the pain, any more than a red curtain (i.e. being of the curtain) is an identifying property of the red colour it has (see §3.8). It is, in our view, a misconception to think that experiences exist only from a firstperson point of view. It is trivially true that A's headache exists only as long as A's head aches - which is a misleading way of saying that A has a headache only if his head aches. But that tautology gives no support to the thought that a person's headache exists only from his point of view. From A's point of view, it may be better to do X rather than Y. From a moral point of view, it may be better to do Y rather than X. And from an economic point of view it may be better to do nothing at all. Differently, X may be visible only from certain points of view or viewpoints — but then anyone can see it from that point in space if it is vacated.22 But experiences are not visible objects that can be seen from certain positions and not from others, although A's fury may be perfectly visible, and B's cheerfulness may be manifest. However, there is no such thing as something's existing only from some person's point of view (let alone from a moral, political or economic point of view) or from some viewpoint that a person may occupy. It is a further error to suppose that the subject of an experience has 'access' to it, that my pain 'is accessible to me in a way that it is not accessible to you'. 23 For to have a pain, as we have pointed out, is not to have access to anything (§3.7). What is true is that if I have a pain, it does not follow that you will have one too, let alone that you will have the same pain. But that humble truth does not imply that you cannot have the very same pain.

22

This does not apply to all the perceptual modalities. Things are not felt to be hot or cold, i dry, smooth or rough, from any point of view or viewpoint. Nor do they have such-and-i taste from any 'viewpoint'. 23 Searle, Mystery of Consciousness, p. 8.

448

Appendix 2; John Searle

as I, such as a splitting headache as a result of drinking too much at yesterday's party, I n Searle's view, the accessibility of a pain to the sufferer 'has epistemic consequences - you can know about your pain in a way that others cannot'. 24 This is, in our view, misconceived. To have a pain is not a way of knowing anything. What is true is that the person who is in pain can say so, whereas others may not be able to say whether he is or is not. This is not because he has access to something to which others lack access. It is because he is in pain, but if he is not showing it, others will not be able to tell that he is. It is mistaken to think that 'conscious states are qualitative in the sense that for any conscious state, such as feeling a pain or worrying about the economic situation, there is something that it qualitatively feels like to be in that state'.25 Of course, there is, in a sense, a 'qualitative feel' to pains. One has an affective attitude towards them. They are, to say the least, very unpleasant to endure, and one would normally prefer not to have any. So there is an answer to the question 'What does it feel like to have migraine?': namely, 'Dreadful' or 'Extraordinarily unpleasant'. But, as we have argued (§§10.3-10.3.5), it would be quite wrong to suppose that every experience, let alone every thought or belief, is distinguished by some special feeling and is the subject of some attitudinal predicate. And it is misconceived to characterize 'all conscious phenomena' as 'qualia', since, as we have argued at length, the notion of a quale, of there being something which it is like to experience this or that, is not coherent. Searle cleaves to what he calls 'The Principle of the Independence of Consciousness and Behaviour', according to which there is no 'conceptual or logical connection between conscious mental phenomena and external behaviour'. 26 The grounds for this principle are that we can imagine circumstances in which appropriate behaviour occurs - for example, pain-behaviour — but without the subject being in pain; and, conversely, we can imagine a person in pain, but without exhibiting any pain-behaviour (e.g. if he is paralysed, or if . the pain is trivial). So behaviour is neither necessary nor sufficient for the presence of the relevant mental phenomenon. We agree that, for the most part, behaviour is neither necessary nor sufficient for a wide array of mental phenomena - pretence and deception are sometimes possible, and concealment, suppression and paralysis are also conceivable. But it does not follow that the mental is not conceptually connected to its behavioural manifestations, or, conversely, that the relevant behavioural manifestations are not conceptually bound up with the mental phenomenon they manifest. And it is incorrect to say that 'the phenomena in question can exist completely and have all of their essential properties independent of any behavioural output'. 27 What is possible some of the time may not (and in this case is not) possible all of the time. There is a conceptual link between inner and outer. Behaviour, in appropriate circumstances, is a logical criterion of the mental. Pretence is not always possible (it makes no sense to suppose that a neonate might be pretending); concealment and suppression of outward manifestations are not always

Ibid., p. 98. Ibid., p. xiv. Searle, Rediscovery of the Mind, p. 69. Ibid.

Appendix 2: John Searle

449

options. And what mental phenomena can intelligibly be ascribed to a creature depends upon what that creature can in principle express within the constraints of its behavioural repertoire. We have argued this at length, both in chapter 3 and elsewhere, and shall not repeat the arguments.

3 The Traditional Mind-Body Problem Searle contends that the traditional mind-body problem is amenable to scientific solution. 28 If that were so, it might indeed vindicate Searle's conception of the relationship between philosophy and science. Philosophers have indeed worried over the relationship between the mind and the body since the dawn of the subject. They have wondered what kind of entity the psuche, mind or soul is: whether it is the 'form' of the body, a principle of life, an immaterial substance, a bundle of experiences causally related to the body, or just the brain itself. They have been baffled by the dilemma of accepting the intelligibility of a causal connection between an immaterial mind and the material brain or denying its intelligibility and thereby denying any causal interaction. They have struggled with the question of the relation between mental states and states of the brain. And so on. Evidently Searle holds that philosophy has now progressed to the point at which it can formulate a sharp question and hand it over to neuroscience for empirical resolution. So, what is held to be the form that the 'mind—body problem' takes that is amenable to scientific solution? It breaks down into two questions: H o w exactly do neurobiological processes in the brain cause conscious states and processes? and H o w exactly are those conscious states and processes realized in the brain?29 Searle admits that these questions look like empirical scientific ones. The task of philosophy is to clear away some obstacles to tackling them, obstacles such as the allegedly obsolete categories of mind and body, matter and spirit, mental and physical; or the thought that since science deals with objective phenomena, and consciousness is subjective, science cannot investigate consciousness; or the worry about the indirectness of the verifying procedures for hypotheses about consciousness. We do not think that questions about the neurobiological causes of mental states are any part of the philosophical problems concerning the mind and the body. How and why imbibing excessive alcohol causes one to be first jocose, then bellicose, later morose and finally comatose can be of no concern to philosophy, and is not a sharpening of any philosophical question. Nor can an investigation of the neural antecedents and concomitants

Searle, 'Future of philosophy', p. 2073. Ibid.

450

Appendix 2: John Searle

of voluntary movement resolve any philosophical questions about freedom of the will or voluntary action. There are, as we have seen, deep philosophical questions about the nature of self-consciousness, but, as we showed in chapter 12, they are not to be solved or resolved by neuroscience. And if one wants to discover whatever neural features make self-consciousness possible, one would search not for neural self-scanning mechanisms in the brain, but for the neural conditions for mastery and use of the psychological vocabulary in the first person. Neuroscience is still very far from achieving any such thing, but if it is ever able to investigate such matters, it seems to us unlikely that it will find anything distinctive about the neural correlates of the use of psychological verbs in the first person as opposed to their third-person employment. The solution to the traditional puzzles about self-consciousness is an analytical, not a neuroscientific, task. We do not think that there is anything obsolete about the categories of mind and body, mind and matter, or the mental and the physical. What is true is that there is much confusion about these concepts. They are commonly misunderstood and misused in philosophical, psychological and cognitive-neuroscientific debate. This does not show obsolescence, but a pressing need for philosophical clarification. It is not as if these concepts should be abandoned, as the concept of phlogiston was jettisoned in chemistry, élan vital in biology, and a ether in cosmology. There is no reason why we should cease talking of having minds of our own, of making up our minds, changing our minds, and having something in mind. Nor is it either likely or necessary that we cease to speak of having a body, cease being proud or ashamed of the body we have, or cease to admire the beautiful bodies of the young. There is nothing awry with the category of matter, although it is true that its customary juxtaposition with the mind and with the category of the mental is typically confused. There is nothing wrong with the general concepts of the physical and the mental, although, again, it is perfectly true that these are vague categories that are .neither mutually exclusive nor conjunctively exhaustive {see Preliminaries to Part II, §3). It is true that they are not very useful for scientific purposes, but they are none the worse for that. For they have perfecdy decent common or garden uses. Searle contends that 'In the case of humans, unless we perform experiments upon ourselves, individually, our only conclusive evidence for the presence and nature of consciousness is what the subject says and does, and subjects are notoriously unreliable.' 30 However, he contends that this 'is no more an obstacle in principle [to a science of consciousness] than the difficulties encountered in other forms of scientific investigation where we have to rely on indirect means of verifying our claims'. We agree that the evidence for the ascription of states of consciousness (and much else that is broadly speaking 'psychological') to other human beings consists of what people do and say. But three important points should be noted. First, such evidence may be logical - as in the case of criterial support - or inductive. There is nothing indirect about criterial evidence: pain-behaviour or conative behaviour, for example, are not established as criterial evidence for pain and desire, respectively, as a result of an inductive correlation. It is correct that criterial evidence is defeasible in certain

Ibid., p. 2074.

Appendix 2: John Searle

451

circumstances (though not all). But if it is not defeated, it is often (though not always) adequate to confer certainty. Secondly, it would be altogether mistaken to suppose that what we observe are mere physical movements, or to think that what we hear are mere sounds, as opposed to intelligible human speech. What we observe are human actions and reactions - chortles of amusement, gestures of rage, cries of joy or grief. When we observe someone behaving thus, there is nothing indirect about the verifying evidence for our immediate judgement that the person is amused, angry, joyous or grieving (although, of course, in certain circumstances deception is possible). Similarly, if someone tells us what he hopes or fears, there is nothing indirect about our consequent knowledge. Indirect knowledge in such a case would be hearsay, by contrast with hearing from the horse's mouth, as it were. Thirdly, although we often justify our ascription of mental states to others by reference to what they do and say, it does not follow (nor does Searle think it does) that our knowledge is always or even typically inferential. On the contrary, we commonly recognize immediately, without any inference, that someone is in agony or overjoyed or tormented by grief (see §§3.3—3.5). Similarly, if someone tells us what he thinks, we do not inferwhzt he thinks from what he said. Consequently we agree with Searle that knowledge of the experiences of other people is not (normally) inferential, a fortiori not analogical. However, we hold it mistaken to suggest that the indirect means of verifying claims concerning black holes or atomic or subatomic particles 'should give us a model for verifying hypotheses in the area of the study of human and animal subjectivity'.31 The states of consciousness of others, their beliefs and thoughts, their hopes and fears, are not the slightest akin, from a logical or epistemological point of view, to black holes or to atomic particles that can be observed only indirectly. To see someone writhing in agony, dancing with joy, or distraught with grief is not to observe their pain, joy or grief indirectly, any more than to hear from their own mouth what they think or believe is to come to know their thoughts and beliefs indirectly. We are in agreement with Searle that it is confused to suppose that because consciousness is 'subjective' and science is 'objective', therefore science cannot investigate the neural basis of conscious mental states. But we do not think that what neuroscience may thus investigate goes any way towards solving or resolving philosophical problems pertaining to consciousness. It should be obvious from antecedent discussions (see Chapter 3) that such a question as 'How is the mind related to the body?' is primarily a philosophical question that needs conceptual, not empirical, investigation. But it is far from clear that the question makes good sense (and this is something which philosophical investigation must elucidate). It requires clarification of the concept of a mind, as a condition of clarifying whether the question of how the mind and the body are related makes any sense. So, for example, if, as we argued (§3.10), discourse about the mind is characteristically zfagon de parler, then it makes no sense to talk of a relation between mind and body. For 'the mind' is not a kind of entity that can stand in relationship to anything.

[bid., p. 2074.

452

Appendix 2: John Searle

So, too, it is a conceptual question how the concepts of a mind and of a person's body are related to the concept of a person. We say that a person has a mind and has a body. Are there three different entities here - a person, a mind and a body? O r two - a mind and a body? Or only one — a person? What is it to have a body? Might a person cease to have a body, but retain his mind? Or vice versa? These are not scientific questions, and are not soluble by scientific means. They are conceptual questions, requiring an overview and description of a field of concepts. There are numerous other philosophical questions in this domain, many of which have been touched upon in the course of this book. We see no reason to suppose that these kinds of question are ever likely to be answered by neuroscience. They are conceptual questions, not empirical ones, and they can be answered only by conceptual investigations. But we agree with Searle that many of these questions need to be answered antecedently to fruitful investigations in the neurosciences, if only to ensure that neuroscientists do not get enmeshed in webs of grammar and misconstrue the questions they are raising and misinterpret the results of their own experiments. That was one reason for writing this book.

w

Index

ability (see also power), 149-51, 157, 164, 170, 267 dawning of, 309n. abstract entities, 50f. access, direct/indirect, 9 2 - 4 access, privileged, see privileged access Ackrill, J. L„ 14n. act of will, see will, act of action explanation of, 3, 64, 227f, 230, 232, 360, 362-6, 365£, 371f, 375 involuntary, 221n., 224f, 229 voluntary, 31f„ 52f., 221n., 224f, 226f, 229f, 235 active intellect, 24n. acts, volitional, see will, act of actuality (entelechiai), 14, 19 Adrian, E. D., 2, 47-9, 233 affections, 199f., 202 agitations, 201f., 220, 221 Albright, T. D., 239 analyticity, 438n. animal spirits (see also psychic pneuma), 27f., 30, 31, 47 animals, 150f., 311-14, 335-7, 424 and concept possession, 340f. consciousness of, 312-14, 317, 321f., 328, 335, 346f. desires of, 336 emotions of, 205£, 208f, 210, 336f. thought of, 336, 345 appetites, 200f. Aquinas, 24, 46, 443

Aristotle, 11-19, 22, 24, 26, 30, 35, 43f., 52, 71n., 159, 233, 325, 379 attention, 91£, 248, 249, 253, 254, 257, 258f. distraction of, 121n., 312n. attitude, 276, 277f. Augustine, 46 Austin, J. L., 442 Avicenna, 22 awareness, 248, 253, 256, 323 problem of, 314-16 Baars, B. j . , 262, 317, 327n. Bain, A., 53 Barlow, H., 142f, 307f, 312f bat, 272, 279f Beck, A., 41 behaviourism, 82n., 117 logical, 413n., 415, 416f. belief, 66, 75, 81f, 93, 94, 172-4, 175, 181, 206f, 218, 268n., 361f., 415, 417, 422, 425, 426, 430f acquisition of concept of, 102f. unconscious, 77f., 268f. Bell, C , 3 5 - 7 Bennett, M. R., 4, 20, 35, 163 Bentham, J., 226 Benton, A. L., 22n. Betts, G. H-, 194 binding problem, 17, 29, 54, 56, 137-43 Blakemore, C , 69, 75, 7 8 - 8 1 , 87, 93f, 136, 152, 182f, 314f, 329£, 331, 333, 356, 379, 3 8 6 - 8 blindsight, 17, 55n., 3 9 3 - 6

454

Index

Block, N . , 271, 274 Boyle, R., 128, 289, 295, 406, 407, 442 brain, 3, 20f., 22£, 29, 32, 33, 34£, 38-40, 41£, 51£, 54£, 58, 59, 60£, 68-70, 72f., 75, 78f., 83£, 88, 112, 122, 139-43, 146, 153, 159, 162-4, 167£, 270£, 356, 359£, 361£, 431£, 433 brain/body dualism, 111-14 Brentano, F„ 420, 421£, 430, 435 Broca, P., 38 Bruner, J., 386 Cajal, Ramón y, 57 caring, 216, 218 Carnap, R., 357n., 438n. Carroü, L., 182n. Cartesianism, 25-30, 32, 46£, 48£, 62, 66£, 68, 72, 85, 103£, 111-14, 233, 294, 312, 317, 382£ Cartesianism and contemporary neuroscience, see neuroscience, crypto-Cartesianism of Catón, R., 40£ certainty, 173, 316, 318 Chalmers, D „ 240, 242, 263, 271, 273, 276, 282, 283, 294, 301, 307, 436 change, 13 child, acquisition of psychological concepts by, 100, 101-3, 347-9, 369f. -Chomsky, N . , 104, 358 Churchland, P. M , 119n., 232n., 366f-, 368, 372, 373, 415 Churchiand, P. S., 119n., 232n., 366, 367, 3 8 0 - 6 , 415 Cicero, 19 classification, 368£ code, 167 coffee, aroma of, 273, 285, 2 8 6 - 8 Cohen, N . , 155 Collins, A. W., 174n., 362n. colour, 5, 96n., 99, 128-35, 282-4, 285, 288, 290f, 399, 406f. commissurotomy, 50, 53£, 56 misdescription of the results of, 78, 106, 153, 388-93 computer, 59, 65, 361n., 431f, 435 conceivability, 383f. concepts, 5£, 60£, 65f, 290, 337, 338f, 3 3 9 - 4 1 , 345, 368-70

possession of, 313n., 340f, 347, 441n. theoretical, 368-70, 374£ confabulation, 389, 391f connective analysis, 378, 399n., 400£, 438 conscious experience, see experience, conscious conscious mental states, see mental states, conscious consciousness, 26, 38, 48f, 86f, 239-71, 293-322, 326, 334f, 337, 397, 403, 431-5, 443-5 of actions, 252, 259£ affective, 250£, 254, 259 of animals, see animals, consciousness of Cartesian conception of, 261£ causation of, 444 etymology of, 256n. evolutionary emergence of, 303f. evolutionary value of, 307—14 and experience, 261-5, 394 feeling of, 241n. intransitive, 2 4 4 - 8 ; causes of, 246 kinaesthetic, 250 and knowledge, 253, 255-7, 257-9, 316 of motives, 251 mystery of, 241f, 243£, 304f. and objective reality, 294—302 perceptual, 249f., 254, 265, 3Í2 and physical processes, 297-300, 302-7 reflective, 251f., 254, 259 somatic, 250, 254, 258£, 312 states of, 246 transitive, 2 4 8 - 6 0 , 263, 318£, 334f. Cooper, L. A., 196 cortex, motor, 39£, 41f. cortical doctrine, 30 Craik, K., 330 Crick, R , 4, 55n., 57, 68, 75, 85n., 87, 127, 138, 140, 144n., 152, 153, 173, 241, 263, 284, 291, 316, 317, 355f., 358n., 359f., 3 8 8 - 9 3 , 397, 398, 403f, 405, 436 criteria, 81-4, 148, 186, 246, 333, 335, 347Í-, 360f, 375f., 448£, 450 Damasio, A., 4, 69, 85, 87n., 89, 90, 138, 166n., 202, 210-16, 234, 241, 271, 275, 303, 305n., 328, 329, 331, 333, 337 Davies, M., 273

Index Dawkins, R., 373, 433 defeasibility, 82f. Democritus, 356 Dennett, D„ 240, 242, 247, 297, 413-35, 436 Descartes, R „ 15, 25-30, 34, 35, 44, 46, 48f., 52, 54, 62, 65n., 68, 103, 113, 115, 128, 226, 233£, 234f., 242, 261, 289, 295, 306, 312, 317, 324, 326, 330, 331, 332n., 358n., 382, 386, 404, 406, 407, 416, 431, 442 description, 94n., 143f., 145£, 167, 284£, 286-9 of qualities, 287f. desire, 200f., 222, 230 unconscious, 77f. Diderot, D., 355n. disposition, 118, 157, 266, 267 dissociation of capacities, 65 dissociation of dysfunctions, 394 Doty, R . W-, 106, 391, 393 dream, 247, 264 du Petit, F. P., 33 dualism, 22, 25-30, 4 3 - 7 , 48f, 51-7, 5 9 - 6 7 , 72, 111-14, 324£, 382f, 409 Dupré, J., 357n. duration, genuine, 266f Eccles, J., 2, 3, 28, 49-57, 62, 72, 233, 409 Edelman, G., 4, 68, 85, 87, 138, 179f, 239, 241, 263, 272, 273, 274, 275, 281, 284, 303, 316, 327n., 328, 337, 340, 346, 348, 397, 398, 402, 403f, 405, 436 Ego, the, 47, 48f, 324f., 332, 347 Einstein, A., 4, 338, 344, 434 eliminative materialism, see materialism, eliminative Elizabeth of Bohemia, 47 emergent properties, 359f. emotion, 199-223, 234, 337 of animals, see animals, emotions of causes of, 220f. duration of, 205 and motives, 204, 222f. objects of, 206, 208, 216-18, 220f. perceptibility of, 22If. reasons for, 213, 218f. somatic accompaniments of, 209f, 213, 214, 221

455

somatic marker hypothesis, 212-16 words, learning of, 213 emotional attitudes, 203f., 213 emotional character traits, 203 emotional perturbations, 202, 203-5, 207, 220 encoding, 160, 164, 167, 169 English, alleged inadequacy of, 379, 387f. engrain, see memory trace epilepsy, Penfield's work on, 57f. epileptic automatism, 59f, 65 epiphenomenalism, 310 Escher, M., 182n. evidence, direct/indirect, 93 experience, 88, 264f, 2 7 4 - 8 4 , 289f. conscious, 263-7, 272f., 2 7 4 - 6 , 277f., 281-92, 297, 303, 333, 394 content of, 282f description of, 284f., 286 epistemic privacy of, 86f, 187n., 241, 243, 246, 286, 295, 448 indescribability of, 286-92 names of, lOOf. private 'ownership' of, 84, 85f, 88, 9 4 - 7 , 241, 286, 295f, 395, 446f. subjective character of, see qualia unity of, 54f, 56 fafon de parkr, 3£, 62£, 75, 104f, 127, 358, 379, 425, 451 factivity, 255£ fantasia, 183£, 185f, 187-98 feeling, 89f, 119, 199f, 203, 208£, 210, 211£, 214, 215, 230, 250£, 274, 275, 276, 281 Fernel, J., 2 3 - 5 , 28, 43 Ferrier, D., 40, 41 Finke, R . A., 188, 196, 198 first-person pronoun, 325, 330, 332, 334, 347£ Flourens, M.-J.-P., 38 Foerster, O., 57 folk psychology, 232, 366-77, 415, 416-18 forgetting, 149, 157, 158 form/matter distinction, 13£, 15 Foster, M., 38 Frege, G., 50, 86 Freud, S., 77 Frisby, J. P., 69, 77, 144n„ 242 Frith, C. D-, 231 Fritsch, G„ 38£

456

Index

functional localization, 16f., 2 0 - 2 , 30, 33, 3 8 - 4 0 , 41f., 58 functional dissociation, 106, 391f. Galen, 11, 12, 19-21, 23, 30, 34f. Galileo, 128, 233, 289, 295, 380, 383, 406, 407 Galton, F., 187 188, 191, 192, 195, 338, 344 Galvani, L., 35 Gassendi, G., 355n. Gazzaniga, M., 57, 78, 106, 153, 160, 164, 240, 388-93 genes, compared with memes, 433f. Gibson, W . G., 163 Glock, H.-J., 380, 438n. Glynn, I., 136, 162, 239, 241, 271, 273, 285, 286, 288, 303, 397, 402 Gosling, J., 228n. Gray, C., 141 Greenfield, S., 70n., 239 Gregory, R., 4, 69, 75, 136, 241 Grice, H. P., 380, 438n. Grünbaum, A. S. F., 41 Guainerio, A., 22 Hacker, P. M. S., 71n., 93n., 97n., 134n., 136n., 146n., 153n., 228n., 231n., 328n., 366n., 380n., 386n., 422n., 438n. H a d a m a r d J . , 309n., 338, 344 Hall, M-, 37, 41 Harvey, W., 25 hearing, 130, 138, 250, 274 Helmholtz, H. von, 53, 135-7 hemispherectomy, see commissurotomy hemispheres of the brain, as subjects of psychological attributes, 78, 106, 388-93 Hertz, H., 5 heterophenomenological method, 427-31 'highest brain mechanism', 5, 59f, 6 1 , 65 Hippocratic tradition, 16n. Hitzig, E., 38f. Hobbes, T., 226, 338n., 355n. Holbach, P.-H. T. d', 355n. homunculus fallacy (see also mereological fallacy), 73n. Horsley, V., 40

human being/person, 3, 16, 26, 29, 45, 6 1 , 63f, 65, 70, 71f, 73f, 83f., 112, 122f., 137, 2 7 9 - 8 1 , 318, 330£, 332, 333, 350f., 356, 359f., 363f., 421, 422f, 425, 451f Hume, D., 193, 211, 226, 326, 443 Humphrey, N . , 86, 87n„ 90, 302, 303, 308, 313£, 329, 331, 333 Huxley, T. H., 302, 303 hydrodynamical analogy, 75, 77 Hyman, j . , 192n., 392n., 395n., 396 hypothesis, 136f., 380, 381 T , the, 44, 47, 48f, 324£, 327, 329f, 331, 332, 333, 347 idea, 433f, 435 identity, numerical/qualitative, 86, 95, 96n. ideo-motor theory of action, 53, 231 image, mental, see mental image image, visual, see visual image imaginability, 383f. imagination, 180-98, 214 'in', different locative uses of, 123 inference, unconscious, 135f., 137 information, 75, 139f, 141, 144f., 146f, 148, 149 storage, 152f., 164, 165 transmitters, 146f. inner/outer, 84f, 88-90 intention, 53, 103, 269 intentional objects, 429, 430 intentional stance, 4 1 9 - 2 1 , 4 2 4 - 7 intentional systems, 421, 426 intentionality, 209, 420, 421-3 interpretation, 80, 420, 426, 427, 429 introspection, 48, 85, 86, 88, 9 0 - 2 , 316, 323, 325f, 351, 395, 419 Ivry, R . B., 160, 164 Jackson, J. H., 5, 39f, 78 James, H., 211n. James, W., 86, 160f, 162, 163, 210f, 231, 232, 326f., 329, 332, 396, 400 james-Lange theory of emotions, 203, 210f., 234 Jessell, T. M „ 113, 129, 132, 239 Johnson-Laird, P. N . , 70, 90, 240, 314, 330 Joynt, R., 22n.

Index

Kandel, E. R., 4, 76, 113, 129, 132, 139£, 154, 155, 159, 162, 163, 239 Kenny, A. J. P., 24n., 71n., 73n., 183n., 186n., 210n., 223n., 228n., 261n., 361n., 363n. kinaesthetic sensation, see sensation knowing how and being able to, 149-51 knowing that, 149, 155n. knowledge, 148-53, 157, 172f., 206f., 218, 253, 269, 406 and ability, 148f. acquisition of, 75, 151f., 166 containment of, 152f. direct/indirect, 93f Hmits of, 335f. possession of, 152f. reception of, 152f., 253, 257-9 retention of, 154, 157, 158, 159, 164, 165, 170 and state, 148f. storage of, 158, 159 of word meaning, 339 Koch, C , 241 Kóhler, W., 161, 162 Kosslyn, S. M., 85, 187, 188 La Mettrie, J. O. de, 355n. language, ordinary, 380-2, 384, 385£, 387f-, 400-2 language-users, 205£, 208£, 232, 313f., 3 3 4 - 7 , 337-46, 3 4 6 - 5 1 , 375£, 418 LeDoux, j „ 152, 158, 159, 207-10, 211, 234 Leonardo da Vinci, 23 Lewis, C. S., 256n. liaison brain, 49, 54£, 56, 59 Libet, B., 52n., 57, 69, 86£, 113n., 2 2 8 - 3 1 , 235, 262 'time-on' theory, 124, 125 linguistic inertia, 381, 384, 385f. Llinás, R., 240 Locke, J., 113, 128, 135n., 226, 233, 289, 295, 325, 326, 329, 331, 338n., 406, 407 logical possibility, 116, 383£, 401 Lomand, E., 272 Luciani, L., 394, 395, 396 Lucretius, 356 Luria, A. R., 189f.

457

Mach, E., 53 Magendie, F., 36f. Malcolm, N., 121, 155n., 165n. Mangun, G. R., 160, 164 maps in the brain, 76, 77, 7 8 - 8 1 , 387f. Marks, D. F., 194 Marr, D., 4, 70, 76, 143-7 materialism eliminative, 366-77 metaphysical, 355n., 357£ ontological, 357-9 mathematics, contrasted with philosophy, 402, 404, 438 McGinn, C , 240 meaning, 339 change of, 384, 385f. of psychological words, 97-103 memes, 416, 431-5 memorizing circuit, 163£ memory, 60, 76, 154-71, 233f. declarative, 76, 155f. experiential, 154, 156, 158 factual, 154, 158 habit, 155£ non-declarative, 155f., 157, 158 objects of, 154, 158, 165f., 170 objectual, 154, 156 storage, 156, 158-71 trace, 160-4, 167, Í68, 169f. memory how/memory that, 155, 156, 157 mental events, 415, 428 mental image, 159f, 160, 166n., 168, 180f., 182-98, 211, 214, 234, 329, 337, 339, 344 and perception, 28f, 137-43, 187-95 rotation of, 195-8 scanning of, 186f. and thinking, 338f., 341£, 343£, 345f. vivacity of, 187, 193-5, 211 mental model, see model, mental mental representations, Í16, 168 mental states, 93n., 112, 268-70, 360f., 381 conscious, 262, 263-7, 273, 297 location of, 361 mereological fallacy, 29, 6 8 - 8 5 , lllf., 114, 137, 144f, Í46, 162n., 379, 381-5, 3 8 6 - 8 , 416

458

Index

mereological fallacy {cont'd) in Descartes, 29 regarding belief, 68, 173f. regarding commissurotomy, 78, 106, 153, 389-93 regarding consciousness and unconsciousness, 86f., 2 3 9 - 4 1 , 246, 270f., 298, 444f. regarding imagination, 182f. regarding knowledge, 152f regarding memory, 154 regarding perception, 127f, 144f. regarding thinking, 179f. regarding volition, 228-31 in Sherrington, 45 mereoiogy, 73 metaphor, 79£, 9 1 , 379, 386f. methodology, 7 4 - 8 1 , 378-409 Metzler, j . , 195£ Meudall, P. R., 190 Miller, M., 390 Milner, B., 75£, 154, 155, 159 mind, 2, 19, 52, 56, 6 3 - 6 , 103-6, 118, 358f, 416, 418, 426f, 433, 450 Cartesian conception of, 2 5 - 7 , 103f Penfield's conception of, 5 9 - 6 7 possession of, 6 3 , 118 Sherrington's conception of, 4 4 - 7 mind/body interaction, 26, 28f., 4 3 - 7 , 113f, '" 325, 449-52 mind/brain relationship/identity, 2f., 47-9, 51£, 53f, 56, 59, 60, 61, 64, 105, 174, 358, 360-3, 380, 416, 430f. Misfichelli, D., 33 mnemonic experience, 169. 170 mnemonic imagery, 154f, 166n., 168 model, mental, 329, 330f., 356 model building, 313f, 329£, 348f. monitoring, 394f neural 55n., 91 moods, 202, 422 More, H., 317n. motive, 202, 204, 222f., 251, 349f. unconscious, 77f. movement involuntary, 28 voluntary, 52f, 225£, 228-31 Mozart, W . A., 308, 309n. Müller, J., 38

multiple drafts model, 43lf. mystery, 242£, 306 Nagel, T., 118, 272f., 281, 284 Nemesius, 12, 21 £, 23, 29 neuroscience (cognitive), 1—7, 111-17, 396-409 crypto-Cartesianism of, 8 5 - 8 , 111-14, 2 3 3 - 5 , 262, 295 and epistemology, 398f-, 406f. and explanation of action, 3, 360, 364—6 history of, 11-42 and philosophy, 396—409 province of, 1, 3, 364£, 367, 405-9, 449£ and psychology, 114f. Newton, I., 4f, 133 nonsense, 74, 78, 133, 242 noticing, 189£, 253 objective, realm of the, 295 objective viewpoint, see viewpoint, objective Ochsner, K. N., 85, 187, 188 Oppenheim, P., 357n. ostensive definition, 97-100, 290£ other minds, knowledge of, 87, 93f, 98£, 311, 316-22, 417£, 451 'outside world', 69n. pain, 8 1 , 82, 88f, 91, 93, 9 4 - 7 , 98f., 121-5, 274, 291£, 312, 319, 320, 333, 347£, 430 acquisition of concept of, 100-2 behaviour, 8 1 , 82, 83, 319 criteria of identity of, 95£, 96£, 291f., 320 expression of, lOlf. location, 95, 122£, 445f. 'ownership' of, 95f. phantom/referred, 123f, 446 subject of, 95, 122£ Penfield, W., 2, 3, 5, 57-67, 72, 233 Penrose, R., 298, 307, 308£, 314, 315n., 338, 344, 345, 351, 436 perception, 16-18, 25, 27, 31£, 9 1 , 125-47, 148, 167, 181, 185, 188£, 188£, 192, 249£, 265, 274, 296, 315f behavioural criteria for, 127 causal account of, 131-4 and cognition, 126, 257f as hypothesis formation, 135—7

Index

as information processing, 143-7 neuron doctrine in, 142f objects of, 129£, 138f. organ of, 125£ peripheral, 258, 312 processes of, 315f. and sensation, 126, 128£, 132f, 274 and voluntariness, 126f. philosophy, 378, 379£, 399-405, 4 3 6 - 4 3 a priori methods of, 397, 402f, 438f, and cognitive neuroscience, 396-409 linguistic turn in, 380 progress in, 404 province of, 1, 6, 402 and science, 4 3 6 - 4 3 and theory, 401, 438-40 and truth, 440f. phobias, 219n. physical, the domain of, 299 physical process and conscious experience, see consciousness and physical processes physical systems, 298-300 physics, 300f, 357, 387 viewpoint of, 294, 301 'Physiologia', 23f. pineal giand, 28£, 30, 32, 52, 54, 325 Plato, 15f, 46, 159 Platonic dualism, 46, 325 Platonism, 50£ pleasure, 275 pneuma, 18£ psychic, 20, 2 1 , 23, 27, 30 vital, 19f. Poincaré, H., 308 point of view, 296£, 415, 416, 418£, 427, 446, 447 poiymorphousness, 178f., 253£ Popper, K., 50 Posner, M. I., 184, 185, 186, 196f, 239 powers, 14, 114, 118£, 126, 149£, 225£, 267 pretence, 246f. primary and secondary qualities, 113, 128-35, 139n., 289 privacy, see experience, epistemic privacy of and experience, private ownership of private language argument, 97-103 private ostensive definition, 98-100, 133, 289, 290

459

privileged access, 84£, 86£, 88, 9 2 - 4 , 96, 243, 247, 295, 333, 394, 395, 447 Procháska, J., 33, 34 progress, 373 prepositional attitudes, 420 propositions, 50£, 420 psuche, 11-19, 20£, 24, 26, 43£ psychological attributes/predicates, 6 8 - 8 8 , 97-104, 111-19, 232, 366, 367-70, 3 7 4 - 6 , 416, 423, 424 psychological concepts, see psychological attributes/predicates psychological laws, 362, 364 psychology, folk, see folk psychology Putnam, H., 357n. qualia, 116, 271-3, 273, 275, 281-92, 295, 297, 300, 305£, 333, 405, 448 incommunicability of, 285f., 316£ qualities, description of, see description of qualities Quine, W. V. O., 380, 413, 414, 436, 438n. Rahman, F., 22 Raichle, M. E., 184, 185, 186, 196f. realization, 249, 253, 257 reasons, 149, 169£, 228 for acting, 149, 314, 363f, 365 recognition, 161, 165, 384£, 1 9 6 - 8 , 254n., 420, 423 recognitional abilities, 319, 341 reductionism, 46£, 302, 355-77 classical, 357 derivational, 357, 360, 362 epiphenomenal, 356 explanatory, 64, 355, 356£, 359, 360, 364 eliminative, 366-77 ontological, 355, 356£, 359 reflection (mirror image), 140n. reflexes, 25, 28, 3 1 , 34, 37, 4 1 , 49 in decorticate animals, 34, 35 Reid, T., 193n. representation, 29, 76, 80, 142£, 160, 164, 165£, 167, 168, 193 internal, 31f, 76, 85, 144n., 147, 192£, 330 representational/non-representational properties, 192£ representationalism, 138, 142£, 144£

460

retinal image, 29, 115, 140, 306, 314 Richardson, j . T. E., 187n. Rio-Hortega, P. del, 57 Robinson, J., 163 Rock, I., 129n. Rolls, E. T., 4, 200 Rosenthal, D. M., 248, 264 Rundle, B., 105, 175n., 363n. Russell, B-, 49, 441 Ryle, G., 151, 155n., 301n., 413, 414 samples, 98, 99£, 289, 290 Savage-Rumbaugh, S., 72 scanning, neural, 55n., 350f., 395f. Schiller, Francis, 49 Schrodinger, E., 43n. Schwartz, J. H., 113, 129, 132 science, 2, 4f., 373f., 401, 404, 419, 437-43 searchlight hypothesis, 55n. Searle, J. R., 85£, 121, 122, 239, 240, 241, 245n., 247, 263, 268f., 271, 272, 274, 275, 276, 294, 295, 310f., 436-52 seeing, 29, 81, 82, 131-5, 136, 138, 141f, 143-7, 274, 275, 282£, 285, 296, 305f., 393-6 'Self of selves', 327 self, the, 47, 3 2 3 - 8 , 334, 337, 338 criteria of identity for, 325, 332n. -' illusion of, 331-4 self-consciousness, 26, 54f, 244, 252, 323-34, 346-51, 450 limits of, 349f. and linguistic ability, 334-7, 349 and self-scanning mechanism, 350f self-knowledge, 350 Sellars, W., 373 semantic inertia, 75, 78 sensation, 45, 73n., 96, 100, 121-5, 128f, 132f, 134, 135, 200f., 395 feeling a, 96f, 121 kinaesthetic, 250 location of, 122f, 445f in the sensorium, 5, 133 sense, bounds of, 6, 116, 399, 401, 439 sense/nonsense, 6, 383, 384, 399, 401f., 407, 439 sense organ, 16, 146, 189 senses, 16-18, 126, 146f.

Index

sensibles, common/proper, 17 sensorium communis, see sensus communis sensus communis, 16-18, 22, 25, 28, 32, 33, 34, 54 Shepard, R., 185, 190, l95f. Sherrington, C , 2, 3, 5, 11, 28, 34, 37, 41f, 4 3 - 7 , 49f., 57, 58, 6 1 , 62, 72, 138, 233 Singer, W., 23n., 54, 141 sleep, 244f somatic-marker hypothesis, see emotion soul, 20£, 22, 24, 25£, 3 l £ immortality of, 22, 24, 32 interaction with body, 2 6 - 7 , 28, 32 soul {Aristotelian conception of), see psuche sound, 129f, 250 speaking, with thought, 342£ Sperry, R . W., 50, 53, 55, 57, 78, 106, 388-93 spinal cord, anterior/posterior roots, 3 5 - 7 spinal soul, 5, 19, 34, 35, 36, 37f, 41£ split-brain, description of results of, see commissurotomy Squire, L. R., 4, 75£, 154, 155, 159, 163 Stitch, S., 366 storage, 159£, 164, 165, 170 Strawson, P. F., 378n., 438n. Stroud, B., 134n. Stuart, A., 33f subjective, realm of the, 295£, 297 subjectivity, 9 4 - 7 , 294-302, 303 substance, 13, 26£, 326 immaterial, 15f, 2 5 - 7 , 46£, 5 4 - 7 , 62-7, 324f, 326, 358 Sutherland, S., 240 symbol, 76, 146, 153n. symbolic descriptions in brain, 69, 76, 77, 144n. Symons, J., 414n. theory/theoretical terms, 367-9, 370-2, 374f., 380n., 381, 382, 385, 401, 404, 438f. thinking, 26, 45, 46, 83, 93, 119, 175-80, 275, 281f., 324, 333f, 3 4 1 - 6 acquisition of concept of, 102f. in images, 214, 343, 3 4 4 - 6 in .language, 341£, 343, 344, 345 thought, 26, 344 of animals, see animal thought and language, 337—46

Index

limits of, 335f, 346 locus of, 179f. Tononi, G., 85, 179f., 239, 241, 272, 273, 274, 275, 281, 337, 340, 397, 403 Turing, A., 330 Tyndall, j . , 302, 306f. Ullman, S„ 74, 76f., 145n. unconscious beliefs, see beliefs, unconscious unconscious desires, see desires, unconscious unconscious motives, see motives, unconscious unconsciousness, 2 4 4 - 6 understanding, 339, 374 in a flash, 309 use, ordinary. 381—3, 384—6 ventricles, 21f-, 23, 25, 27f. ventricular doctrine, 2 1 - 3 , 28, 30 Vesalius, A., 23, 30 viewpoint, objective, 294, 301 viewpoint of physics, see physics, viewpoint of visual images (see also mental image), 137-43, 305 visual sensation/visual perception, 394f.

461

vivacity, 193-5 volition, 2 2 4 - 8 , 235 von Wright, G. H-, 363n. Weiskrantz, L., 55n., 9 1 , 317, 351n., 3 9 3 - 6 White, A. R., 150n., 151n., 182n„ 223n., 228n., 248m, 253n., 363n. Whitehead, A. N., 130 Whytt, R., 34, 35 will, act of, 51-3, 113, 226f. Willis, T., 23, 3 0 - 3 , 34, 35 Wittgenstein, L., 7 1 , 78, 89, 97-103, 105, 228n., 231n., 247, 248, 266n., 282f, 299, 305, 311, 327n., 362, 378, 386, 413, 414, 438n. Wolford, G., 390 Wren, Christopher, 30 Wundt, W., 53, 327n., 332 Wurtz, R., 4, 139f. Young, j . Z., 69, 77, 79, 80, 136, 152, 153 Zeki, S., 70n., 75, 152, 398f, 404, 406f. zombies, 310f.